import pandas as pan
import numpy as np
import plotly.express as px
pays = "en Wallonie-Bruxelles (Belgique)"
media1 = pan.read_csv("belgique-motsSeuls-media-nettoye.csv", low_memory=False, names=["mot","interactions"])
nonmedia1 = pan.read_csv("belgique-motsSeuls-nonmedia-nettoye.csv", low_memory=False, names=["mot","interactions"])
media1
| mot | interactions | |
|---|---|---|
| 0 | vie | 175299 |
| 1 | bienvenue | 175299 |
| 2 | émilie | 175299 |
| 3 | dylan | 175299 |
| 4 | faire | 175299 |
| ... | ... | ... |
| 8426089 | timon | 7 |
| 8426090 | wellenreuther | 7 |
| 8426091 | club | 7 |
| 8426092 | bruxellois | 7 |
| 8426093 | renforcer | 7 |
8426094 rows × 2 columns
nonmedia1
| mot | interactions | |
|---|---|---|
| 0 | moment | 224233 |
| 1 | tendresse | 224233 |
| 2 | bébé | 224233 |
| 3 | bébé | 224233 |
| 4 | apprenon | 224233 |
| ... | ... | ... |
| 23350240 | ici | 5 |
| 23350241 | visitgaume | 5 |
| 23350242 | soleildegaume | 5 |
| 23350243 | voter | 5 |
| 23350244 | logo | 5 |
23350245 rows × 2 columns
media1_table = pan.pivot_table(media1,index=["mot"],values=["mot","interactions"],aggfunc=[len,np.sum])
nonmedia1_table = pan.pivot_table(nonmedia1,index=["mot"],values=["mot","interactions"],aggfunc=[len,np.sum])
media1_table
| len | sum | |
|---|---|---|
| interactions | interactions | |
| mot | ||
| #106cabrel | 1 | 70 |
| #balades | 1 | 45 |
| #breath4life | 1 | 9 |
| #cestbienlecas | 1 | 422 |
| #confinemer | 1 | 49 |
| ... | ... | ... |
| 그ー | 1 | 34 |
| 그切 | 1 | 36 |
| 나4 | 1 | 5 |
| 무를 | 1 | 414 |
| 제이행 | 1 | 2645 |
128697 rows × 2 columns
nonmedia1_table
| len | sum | |
|---|---|---|
| interactions | interactions | |
| mot | ||
| #1 | 1 | 79 |
| #2 | 3 | 24 |
| #2020toussolidaire | 1 | 826 |
| #21juillet | 1 | 693 |
| #25 | 1 | 56 |
| ... | ... | ... |
| single | 2 | 85 |
| solidaire | 1 | 12 |
| watch | 1 | 13 |
| 𐀁мвя | 1 | 12 |
| 𐌾ᗩïᗩᙅ | 1 | 11 |
471368 rows × 2 columns
media1_table["type"] = "media"
nonmedia1_table["type"] = "non-media"
media1_table
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| mot | |||
| #106cabrel | 1 | 70 | media |
| #balades | 1 | 45 | media |
| #breath4life | 1 | 9 | media |
| #cestbienlecas | 1 | 422 | media |
| #confinemer | 1 | 49 | media |
| ... | ... | ... | ... |
| 그ー | 1 | 34 | media |
| 그切 | 1 | 36 | media |
| 나4 | 1 | 5 | media |
| 무를 | 1 | 414 | media |
| 제이행 | 1 | 2645 | media |
128697 rows × 3 columns
nonmedia1_table
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| mot | |||
| #1 | 1 | 79 | non-media |
| #2 | 3 | 24 | non-media |
| #2020toussolidaire | 1 | 826 | non-media |
| #21juillet | 1 | 693 | non-media |
| #25 | 1 | 56 | non-media |
| ... | ... | ... | ... |
| single | 2 | 85 | non-media |
| solidaire | 1 | 12 | non-media |
| watch | 1 | 13 | non-media |
| 𐀁мвя | 1 | 12 | non-media |
| 𐌾ᗩïᗩᙅ | 1 | 11 | non-media |
471368 rows × 3 columns
tableau1 = pan.concat([media1_table,nonmedia1_table], names = ["mot","nb","interactions","type"], levels=0)
tableau1
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| mot | |||
| #106cabrel | 1 | 70 | media |
| #balades | 1 | 45 | media |
| #breath4life | 1 | 9 | media |
| #cestbienlecas | 1 | 422 | media |
| #confinemer | 1 | 49 | media |
| ... | ... | ... | ... |
| single | 2 | 85 | non-media |
| solidaire | 1 | 12 | non-media |
| watch | 1 | 13 | non-media |
| 𐀁мвя | 1 | 12 | non-media |
| 𐌾ᗩïᗩᙅ | 1 | 11 | non-media |
600065 rows × 3 columns
tableau1["mot"] = tableau1.index
tableau1.columns = ["nb","interactions","media","mot"]
tableau1.reset_index(drop=True, inplace=True)
tableau1
| nb | interactions | media | mot | |
|---|---|---|---|---|
| 0 | 1 | 70 | media | #106cabrel |
| 1 | 1 | 45 | media | #balades |
| 2 | 1 | 9 | media | #breath4life |
| 3 | 1 | 422 | media | #cestbienlecas |
| 4 | 1 | 49 | media | #confinemer |
| ... | ... | ... | ... | ... |
| 600060 | 2 | 85 | non-media | single |
| 600061 | 1 | 12 | non-media | solidaire |
| 600062 | 1 | 13 | non-media | watch |
| 600063 | 1 | 12 | non-media | 𐀁мвя |
| 600064 | 1 | 11 | non-media | 𐌾ᗩïᗩᙅ |
600065 rows × 4 columns
khi2_1 = pan.pivot_table(tableau1,index=["mot"],columns=["media"],values=["mot","interactions"],aggfunc=[len,np.sum],fill_value=0)
khi2_1
| len | sum | |||||||
|---|---|---|---|---|---|---|---|---|
| interactions | nb | interactions | nb | |||||
| media | media | non-media | media | non-media | media | non-media | media | non-media |
| mot | ||||||||
| #1 | 0 | 1 | 0 | 1 | 0 | 79 | 0 | 1 |
| #106cabrel | 1 | 0 | 1 | 0 | 70 | 0 | 1 | 0 |
| #2 | 0 | 1 | 0 | 1 | 0 | 24 | 0 | 3 |
| #2020toussolidaire | 0 | 1 | 0 | 1 | 0 | 826 | 0 | 1 |
| #21juillet | 0 | 1 | 0 | 1 | 0 | 693 | 0 | 1 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... |
| single | 0 | 1 | 0 | 1 | 0 | 85 | 0 | 2 |
| solidaire | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| watch | 0 | 1 | 0 | 1 | 0 | 13 | 0 | 1 |
| 𐀁мвя | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| 𐌾ᗩïᗩᙅ | 0 | 1 | 0 | 1 | 0 | 11 | 0 | 1 |
506548 rows × 8 columns
khi2_1["mot"] = khi2_1.index
khi2_1.columns = khi2_1.columns.get_level_values(0)
khi2_1.reset_index(drop=True, inplace=True)
khi2_1.columns = ["a","b","c","d","interactions_media","interactions_nonmedia","obs_media","obs_nonmedia","mot"]
khi2_1 = khi2_1.drop(columns=["a","b","c","d"])
khi2_1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | |
|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 |
| 1 | 70 | 0 | 1 | 0 | #106cabrel |
| 2 | 0 | 24 | 0 | 3 | #2 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire |
| 4 | 0 | 693 | 0 | 1 | #21juillet |
| ... | ... | ... | ... | ... | ... |
| 506543 | 0 | 85 | 0 | 2 | single |
| 506544 | 0 | 12 | 0 | 1 | solidaire |
| 506545 | 0 | 13 | 0 | 1 | watch |
| 506546 | 0 | 12 | 0 | 1 | 𐀁мвя |
| 506547 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ |
506548 rows × 5 columns
khi2_1.obs_media.sum(), khi2_1.obs_nonmedia.sum()
(8426093, 23350157)
print("Nb de lignes fichier media1 = ", media1.shape[0])
print("Somme observée média khi2_1 = ", khi2_1.obs_media.sum())
print("Nb de lignes fichier nonmedia1 = ", nonmedia1.shape[0])
print("Somme observée nonmédia khi2_1 = ", khi2_1.obs_nonmedia.sum())
Nb de lignes fichier media1 = 8426094 Somme observée média khi2_1 = 8426093 Nb de lignes fichier nonmedia1 = 23350245 Somme observée nonmédia khi2_1 = 23350157
khi2_1["exp_media"] = ((khi2_1.obs_media + khi2_1.obs_nonmedia) * khi2_1.obs_media.sum()) / (khi2_1.obs_media.sum() + khi2_1.obs_nonmedia.sum())
khi2_1["exp_nonmedia"] = ((khi2_1.obs_media + khi2_1.obs_nonmedia) * khi2_1.obs_nonmedia.sum()) / (khi2_1.obs_media.sum() + khi2_1.obs_nonmedia.sum())
khi2_1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | |
|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 | 0.265170 | 0.734830 |
| 1 | 70 | 0 | 1 | 0 | #106cabrel | 0.265170 | 0.734830 |
| 2 | 0 | 24 | 0 | 3 | #2 | 0.795509 | 2.204491 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire | 0.265170 | 0.734830 |
| 4 | 0 | 693 | 0 | 1 | #21juillet | 0.265170 | 0.734830 |
| ... | ... | ... | ... | ... | ... | ... | ... |
| 506543 | 0 | 85 | 0 | 2 | single | 0.530339 | 1.469661 |
| 506544 | 0 | 12 | 0 | 1 | solidaire | 0.265170 | 0.734830 |
| 506545 | 0 | 13 | 0 | 1 | watch | 0.265170 | 0.734830 |
| 506546 | 0 | 12 | 0 | 1 | 𐀁мвя | 0.265170 | 0.734830 |
| 506547 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ | 0.265170 | 0.734830 |
506548 rows × 7 columns
khi2_1["res_media"] = (khi2_1.obs_media - khi2_1.exp_media) / np.sqrt(khi2_1.exp_media)
khi2_1["res_nonmedia"] = (khi2_1.obs_nonmedia - khi2_1.exp_nonmedia) / np.sqrt(khi2_1.exp_nonmedia)
khi2_1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
| 1 | 70 | 0 | 1 | 0 | #106cabrel | 0.265170 | 0.734830 | 1.427005 | -0.857223 |
| 2 | 0 | 24 | 0 | 3 | #2 | 0.795509 | 2.204491 | -0.891913 | 0.535785 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
| 4 | 0 | 693 | 0 | 1 | #21juillet | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 506543 | 0 | 85 | 0 | 2 | single | 0.530339 | 1.469661 | -0.728244 | 0.437467 |
| 506544 | 0 | 12 | 0 | 1 | solidaire | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
| 506545 | 0 | 13 | 0 | 1 | watch | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
| 506546 | 0 | 12 | 0 | 1 | 𐀁мвя | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
| 506547 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ | 0.265170 | 0.734830 | -0.514946 | 0.309336 |
506548 rows × 9 columns
graph_media1 = khi2_1.sort_values(by=["res_media"],ascending=False).head(50)
graph_media1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 132650 | 19325018 | 14609115 | 65421 | 40483 | coronavirus | 28082.512980 | 77821.487020 | 222.812080 | -133.846480 |
| 85116 | 11838300 | 9946661 | 39230 | 36723 | belgique | 20140.420648 | 55812.579352 | 134.512328 | -80.803526 |
| 84949 | 7968684 | 6626220 | 29100 | 25891 | belge | 14581.937144 | 40409.062856 | 120.226737 | -72.221962 |
| 61895 | 4996615 | 6088391 | 21922 | 19685 | annoncer | 11032.908271 | 30574.091729 | 103.668461 | -62.275163 |
| 445838 | 2212544 | 1126940 | 10607 | 4776 | trump | 4079.102746 | 11303.897254 | 102.209435 | -61.398705 |
| 60496 | 1647774 | 504751 | 8958 | 3632 | anderlecht | 3338.484273 | 9251.515727 | 97.257751 | -58.424156 |
| 151790 | 3824512 | 613719 | 7715 | 2487 | diable | 2705.259456 | 7496.740544 | 96.318735 | -57.860075 |
| 419647 | 1562305 | 947363 | 9051 | 4036 | standard | 3470.273525 | 9616.726475 | 94.734656 | -56.908496 |
| 90053 | 1594174 | 509468 | 6908 | 2791 | bilan | 2571.879187 | 7127.120813 | 85.501964 | -51.362282 |
| 134732 | 5833072 | 7996479 | 24467 | 29475 | covid-19 | 14303.774316 | 39638.225684 | 84.977949 | -51.047499 |
| 122413 | 2559615 | 4069796 | 17655 | 18057 | club | 9469.733943 | 26242.266057 | 84.113153 | -50.528003 |
| 270878 | 2243930 | 1023084 | 8705 | 5258 | league | 3702.562025 | 10260.437975 | 82.211116 | -49.385421 |
| 157498 | 1427126 | 616647 | 6287 | 2621 | donald | 2362.130095 | 6545.869905 | 80.755732 | -48.511151 |
| 163866 | 2163216 | 1355137 | 8557 | 5302 | décès | 3674.984395 | 10184.015605 | 80.532543 | -48.377078 |
| 117186 | 1866221 | 1391811 | 7945 | 4551 | chiffre | 3313.558338 | 9182.441662 | 80.457936 | -48.332261 |
| 197835 | 1210570 | 1153861 | 8066 | 4841 | football | 3422.543011 | 9484.456989 | 79.371935 | -47.679884 |
| 59873 | 2115405 | 1717005 | 10378 | 8005 | américain | 4874.611309 | 13508.388691 | 78.824312 | -47.350919 |
| 109634 | 4144422 | 5605739 | 19009 | 22028 | cas | 10881.761644 | 30155.238356 | 77.910034 | -46.801698 |
| 325957 | 8847306 | 17145420 | 41255 | 65168 | nouveau | 28220.135961 | 78202.864039 | 77.593788 | -46.611725 |
| 100987 | 764641 | 218762 | 4287 | 1065 | bruges | 1419.187278 | 3932.812722 | 76.125636 | -45.729784 |
| 487072 | 2305015 | 1620938 | 8960 | 6472 | épidémie | 4092.096052 | 11339.903948 | 76.097272 | -45.712746 |
| 282842 | 4801225 | 6327805 | 21076 | 26267 | lundi | 12553.920645 | 34789.079355 | 76.059924 | -45.690310 |
| 130366 | 1246167 | 744332 | 5763 | 2563 | contamination | 2207.801434 | 6118.198566 | 75.663052 | -45.451904 |
| 298109 | 3672515 | 4664767 | 17463 | 19992 | mercredi | 9931.924419 | 27523.075581 | 75.568413 | -45.395053 |
| 103817 | 229005 | 85140 | 4261 | 1112 | bx1 | 1424.755838 | 3948.244162 | 75.140381 | -45.137927 |
| 59979 | 13791596 | 29344381 | 39789 | 64040 | an | 27532.286223 | 76296.713777 | 73.867416 | -44.373239 |
| 341732 | 1797131 | 2704078 | 10524 | 9338 | pandémie | 5266.797031 | 14595.202969 | 72.440502 | -43.516071 |
| 182234 | 1023335 | 526013 | 5174 | 2265 | etats-unis | 1972.596069 | 5466.403931 | 72.081098 | -43.300171 |
| 223337 | 3107469 | 359102 | 3590 | 757 | hazard | 1152.691909 | 3194.308091 | 71.788323 | -43.124297 |
| 251958 | 2522203 | 3467963 | 10756 | 9852 | joueur | 5464.613494 | 15143.386506 | 71.579681 | -42.998963 |
| 323511 | 2340582 | 2677376 | 11777 | 11500 | nombre | 6172.350946 | 17104.649054 | 71.338359 | -42.853997 |
| 114287 | 2222118 | 2288515 | 6824 | 4437 | champion | 2986.073979 | 8274.926021 | 70.233824 | -42.190487 |
| 290644 | 3496090 | 4659470 | 16055 | 19062 | mardi | 9311.958078 | 25805.041922 | 69.877177 | -41.976244 |
| 222806 | 821491 | 711923 | 4247 | 1593 | hausse | 1548.590004 | 4291.409996 | 68.570878 | -41.191531 |
| 302676 | 4377445 | 5782810 | 17955 | 22984 | ministre | 10855.775031 | 30083.224969 | 68.136610 | -40.930660 |
| 249970 | 3958967 | 4938957 | 15846 | 19397 | jeudi | 9345.369438 | 25897.630562 | 67.244574 | -40.394801 |
| 366329 | 3230123 | 3949865 | 16141 | 20202 | président | 9637.055911 | 26705.944089 | 66.252857 | -39.799062 |
| 441269 | 2201255 | 3044567 | 10611 | 10678 | tour | 5645.193938 | 15643.806062 | 66.092217 | -39.702563 |
| 454506 | 1298407 | 1516450 | 6317 | 4304 | vaccin | 2816.365485 | 7804.634515 | 65.963283 | -39.625110 |
| 163879 | 2690518 | 2578659 | 6311 | 4299 | décéder | 2813.448621 | 7796.551379 | 65.939342 | -39.610729 |
| 149042 | 3922699 | 9062212 | 17206 | 22532 | dernier | 10537.306436 | 29200.693564 | 64.964478 | -39.025114 |
| 232582 | 3917914 | 4788672 | 10125 | 10201 | hôpital | 5389.835689 | 14936.164311 | 64.498155 | -38.744986 |
| 428823 | 2051059 | 3653799 | 9189 | 8711 | t | 4746.534431 | 13153.465569 | 64.481592 | -38.735037 |
| 238792 | 844371 | 560880 | 3990 | 1660 | infection | 1498.207795 | 4151.792205 | 64.376268 | -38.671767 |
| 459886 | 520436 | 441857 | 4741 | 2509 | verviers | 1922.479029 | 5327.520971 | 64.282161 | -38.615235 |
| 389330 | 3887735 | 2284497 | 10170 | 10484 | rouge | 5476.811292 | 15177.188708 | 63.416771 | -38.095384 |
| 237347 | 771795 | 691091 | 4411 | 2211 | incendie | 1755.952570 | 4866.047430 | 63.360106 | -38.061344 |
| 457748 | 4694112 | 6621349 | 18947 | 26733 | vendredi | 12112.943731 | 33567.056269 | 62.094581 | -37.301124 |
| 229085 | 704232 | 356016 | 3357 | 1191 | hospitalisation | 1205.990983 | 3342.009017 | 61.939857 | -37.208180 |
| 308194 | 3048268 | 5221458 | 10679 | 11654 | mort | 5922.030918 | 16410.969082 | 61.815159 | -37.133272 |
fig = px.bar(graph_media1, x="res_media", y="mot", labels={"res_media":"Résiduel de Pearson","mot":"Lemme (dans les posts Facebook <b>médias</b>)"}, orientation="h", title="Lemmes les plus caractéristiques des posts Facebook <b>médias</b> {} en 2020".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="red", opacity=0.75, textfont_size=12)
fig
graph_nonmedia1 = khi2_1.sort_values(by=["res_nonmedia"],ascending=False).head(50)
graph_nonmedia1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 358150 | 589111 | 19280588 | 2638 | 62212 | post | 17196.243454 | 47653.756546 | -111.017690 | 66.689952 |
| 201650 | 450691 | 17394426 | 1252 | 53169 | from | 14430.790517 | 39990.209483 | -109.706029 | 65.902018 |
| 350552 | 2568016 | 28083848 | 8714 | 85736 | photo | 25045.261283 | 69404.738717 | -103.194593 | 61.990504 |
| 488547 | 2070626 | 22454212 | 7169 | 51108 | être | 15453.284191 | 42823.715809 | -66.641474 | 40.032509 |
| 22779 | 2049763 | 12657714 | 12851 | 68829 | 2020 | 21659.046497 | 60020.953503 | -59.849418 | 35.952422 |
| 239067 | 2026791 | 7751518 | 5481 | 39147 | info | 11833.985395 | 32794.014605 | -58.399932 | 35.081695 |
| 75786 | 1652840 | 15207630 | 5267 | 37873 | avoir | 11439.413147 | 31700.586853 | -57.710271 | 34.667405 |
| 465870 | 1536650 | 11700589 | 5609 | 38487 | vous | 11692.915210 | 32403.084790 | -56.262853 | 33.797920 |
| 340701 | 416861 | 14410454 | 2132 | 23092 | page | 6688.636004 | 18535.363996 | -55.715439 | 33.469080 |
| 435543 | 570475 | 9326705 | 2785 | 26021 | the | 7638.473229 | 21167.526771 | -55.532773 | 33.359350 |
| 237026 | 419335 | 6675426 | 1569 | 19993 | in | 5717.585217 | 15844.414783 | -54.864786 | 32.958081 |
| 154887 | 273251 | 5486574 | 1786 | 20071 | disponible | 5795.810226 | 16061.189774 | -52.670415 | 31.639890 |
| 28573 | 89325 | 11357803 | 338 | 12178 | 3 | 3318.861728 | 9197.138272 | -51.742481 | 31.082466 |
| 56512 | 5433 | 5089805 | 16 | 9989 | allah | 2653.021060 | 7351.978940 | -51.196851 | 30.754698 |
| 330471 | 169051 | 3002291 | 2015 | 20258 | of | 5906.120747 | 16366.879253 | -50.631821 | 30.415276 |
| 94785 | 3274888 | 34801436 | 14457 | 68346 | bon | 21956.831869 | 60846.168131 | -50.613537 | 30.404293 |
| 152655 | 107371 | 7476766 | 386 | 10792 | dieu | 2964.064909 | 8213.935091 | -47.353272 | 28.445804 |
| 277735 | 298537 | 3803846 | 1266 | 15040 | lire | 4323.854214 | 11982.145786 | -46.503030 | 27.935052 |
| 170172 | 10470 | 1141065 | 36 | 8299 | een | 2210.187960 | 6124.812040 | -46.246885 | 27.781181 |
| 188929 | 6160031 | 44825968 | 25377 | 102298 | faire | 33855.518627 | 93819.481373 | -46.079233 | 27.680470 |
| 430897 | 8481 | 518139 | 47 | 8249 | taquin | 2199.846348 | 6096.153652 | -45.900441 | 27.573067 |
| 276244 | 185924 | 3624725 | 1140 | 14187 | lien | 4064.253252 | 11262.746748 | -45.869563 | 27.554518 |
| 87511 | 1067313 | 7980031 | 2818 | 21415 | besoin | 6425.853009 | 17807.146991 | -45.007350 | 27.036574 |
| 343768 | 866330 | 10256637 | 3083 | 22249 | partager | 6717.274313 | 18614.725687 | -44.342601 | 26.637250 |
| 437949 | 110254 | 2815200 | 755 | 11515 | timeline | 3253.630026 | 9016.369974 | -43.804414 | 26.313953 |
| 244725 | 114856 | 2127421 | 1246 | 13931 | is | 4024.477824 | 11152.522176 | -43.797787 | 26.309972 |
| 240595 | 33678 | 595149 | 405 | 9496 | inscription | 2625.443430 | 7275.556570 | -43.334958 | 26.031944 |
| 233293 | 1293568 | 13928809 | 6202 | 33565 | ici | 10544.996352 | 29222.003648 | -42.292781 | 25.405893 |
| 53548 | 341488 | 8451007 | 1034 | 12027 | aimer | 3463.379117 | 9597.620883 | -41.280531 | 24.797820 |
| 54667 | 112061 | 2891901 | 567 | 9674 | al | 2715.601067 | 7525.398933 | -41.230899 | 24.768005 |
| 225660 | 215204 | 4882694 | 815 | 10841 | het | 3090.815940 | 8565.184060 | -40.935544 | 24.590581 |
| 60387 | 71716 | 2783388 | 484 | 9130 | and | 2549.339777 | 7064.660223 | -40.905118 | 24.572303 |
| 192130 | 2175865 | 12615606 | 6712 | 34608 | femme | 10956.804619 | 30363.195381 | -40.552324 | 24.360375 |
| 438995 | 68438 | 2418437 | 450 | 8799 | to | 2452.552902 | 6796.447098 | -40.436616 | 24.290868 |
| 71463 | 70369 | 733185 | 696 | 10078 | atelier | 2856.936422 | 7917.063578 | -40.428879 | 24.286220 |
| 130257 | 163471 | 2277940 | 680 | 9993 | contacter | 2830.154300 | 7842.845700 | -40.417046 | 24.279112 |
| 349213 | 3116610 | 31943611 | 12426 | 54765 | petit | 17817.005303 | 49373.994697 | -40.388005 | 24.261666 |
| 232509 | 206495 | 2553774 | 1049 | 11757 | hésiter | 3395.760889 | 9410.239111 | -40.271730 | 24.191818 |
| 58920 | 1237266 | 14772671 | 2284 | 17228 | ami | 5173.987699 | 14338.012301 | -40.177536 | 24.135234 |
| 334523 | 5822 | 4162843 | 66 | 6442 | op | 1725.723244 | 4782.276756 | -39.953065 | 24.000391 |
| 163727 | 1436918 | 6975226 | 7932 | 38644 | découvrir | 12350.535622 | 34225.464378 | -39.759001 | 23.883814 |
| 377224 | 125561 | 2227529 | 1132 | 11936 | recette | 3465.235304 | 9602.764696 | -39.636215 | 23.810055 |
| 278196 | 335581 | 3535564 | 1437 | 12985 | live | 3824.274836 | 10597.725164 | -38.603606 | 23.189751 |
| 176891 | 568284 | 5043973 | 2690 | 18287 | ensemble | 5562.461048 | 15414.538952 | -38.514180 | 23.136032 |
| 461195 | 3212547 | 23340821 | 9320 | 42865 | vie | 13837.871467 | 38347.128533 | -38.406013 | 23.071055 |
| 451567 | 24611 | 992869 | 139 | 6354 | une | 1721.745702 | 4771.254298 | -38.144036 | 22.913681 |
| 59399 | 474552 | 12145865 | 1436 | 12786 | amour | 3771.240931 | 10450.759069 | -38.026780 | 22.843244 |
| 393719 | 58761 | 930736 | 527 | 8297 | réservation | 2339.855856 | 6484.144144 | -37.477337 | 22.513185 |
| 353526 | 220403 | 3155573 | 1132 | 11177 | plaisir | 3263.971637 | 9045.028363 | -37.317130 | 22.416947 |
| 465257 | 3389 | 875844 | 48 | 5542 | voor | 1482.297624 | 4107.702376 | -37.253885 | 22.378955 |
fig = px.bar(graph_nonmedia1, x="res_nonmedia", y="mot", labels={"res_nonmedia":"Résiduel de Pearson","mot":"Lemme (dans les posts Facebook <b>non-médias</b>)"}, orientation="h", title="Lemmes les plus caractéristiques des posts Facebook <b>non-médias</b> {} en 2020".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="darkorange", opacity=0.75, textfont_size=12)
fig
# Tentative d'utiliser le logarithme naturel des interactions; donne des résultats ininterprétables
# khi2_1["interactions_media_log"] = np.where(khi2_1.interactions_media == 0, 0, np.log(khi2_1.interactions_media+1e-12))
# khi2_1["interactions_nonmedia_log"] = np.where(khi2_1.interactions_nonmedia == 0, 0, np.log(khi2_1.interactions_nonmedia+1e-12))
# khi2_1
khi2_1["exp_inter_media"] = ((khi2_1.interactions_media + khi2_1.interactions_nonmedia) * khi2_1.interactions_media.sum()) / (khi2_1.interactions_media.sum() + khi2_1.interactions_nonmedia.sum())
# khi2_1["exp_inter_media"] = ((khi2_1.interactions_media_log + khi2_1.interactions_nonmedia_log) * khi2_1.interactions_media_log.sum()) / (khi2_1.interactions_media_log.sum() + khi2_1.interactions_nonmedia_log.sum())
khi2_1["exp_inter_nonmedia"] = ((khi2_1.interactions_media + khi2_1.interactions_nonmedia) * khi2_1.interactions_nonmedia.sum()) / (khi2_1.interactions_media.sum() + khi2_1.interactions_nonmedia.sum())
# khi2_1["exp_inter_nonmedia"] = ((khi2_1.interactions_media_log + khi2_1.interactions_nonmedia_log) * khi2_1.interactions_nonmedia_log.sum()) / (khi2_1.interactions_media_log.sum() + khi2_1.interactions_nonmedia_log.sum())
khi2_1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 15.968818 | 63.031182 |
| 1 | 70 | 0 | 1 | 0 | #106cabrel | 0.265170 | 0.734830 | 1.427005 | -0.857223 | 14.149586 | 55.850414 |
| 2 | 0 | 24 | 0 | 3 | #2 | 0.795509 | 2.204491 | -0.891913 | 0.535785 | 4.851287 | 19.148713 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 166.965111 | 659.034889 |
| 4 | 0 | 693 | 0 | 1 | #21juillet | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 140.080898 | 552.919102 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 506543 | 0 | 85 | 0 | 2 | single | 0.530339 | 1.469661 | -0.728244 | 0.437467 | 17.181640 | 67.818360 |
| 506544 | 0 | 12 | 0 | 1 | solidaire | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.425643 | 9.574357 |
| 506545 | 0 | 13 | 0 | 1 | watch | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.627780 | 10.372220 |
| 506546 | 0 | 12 | 0 | 1 | 𐀁мвя | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.425643 | 9.574357 |
| 506547 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.223506 | 8.776494 |
506548 rows × 11 columns
khi2_1["res_inter_media"] = (khi2_1.interactions_media - khi2_1.exp_inter_media) / np.sqrt(khi2_1.exp_inter_media)
# khi2_1["res_inter_media"] = (khi2_1.interactions_media_log - khi2_1.exp_inter_media) / np.sqrt(khi2_1.exp_inter_media)
khi2_1["res_inter_nonmedia"] = (khi2_1.interactions_nonmedia - khi2_1.exp_inter_nonmedia) / np.sqrt(khi2_1.exp_inter_nonmedia)
# khi2_1["res_inter_nonmedia"] = (khi2_1.interactions_nonmedia_log - khi2_1.exp_inter_nonmedia) / np.sqrt(khi2_1.exp_inter_nonmedia)
khi2_1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 15.968818 | 63.031182 | -3.996100 | 2.011384 |
| 1 | 70 | 0 | 1 | 0 | #106cabrel | 0.265170 | 0.734830 | 1.427005 | -0.857223 | 14.149586 | 55.850414 | 14.847541 | -7.473313 |
| 2 | 0 | 24 | 0 | 3 | #2 | 0.795509 | 2.204491 | -0.891913 | 0.535785 | 4.851287 | 19.148713 | -2.202564 | 1.108631 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 166.965111 | 659.034889 | -12.921498 | 6.503865 |
| 4 | 0 | 693 | 0 | 1 | #21juillet | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 140.080898 | 552.919102 | -11.835578 | 5.957281 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 506543 | 0 | 85 | 0 | 2 | single | 0.530339 | 1.469661 | -0.728244 | 0.437467 | 17.181640 | 67.818360 | -4.145074 | 2.086368 |
| 506544 | 0 | 12 | 0 | 1 | solidaire | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.425643 | 9.574357 | -1.557448 | 0.783921 |
| 506545 | 0 | 13 | 0 | 1 | watch | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.627780 | 10.372220 | -1.621043 | 0.815931 |
| 506546 | 0 | 12 | 0 | 1 | 𐀁мвя | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.425643 | 9.574357 | -1.557448 | 0.783921 |
| 506547 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ | 0.265170 | 0.734830 | -0.514946 | 0.309336 | 2.223506 | 8.776494 | -1.491143 | 0.750547 |
506548 rows × 13 columns
graph_interactions_media1 = khi2_1.sort_values(by=["res_inter_media"],ascending=False).head(50)
graph_interactions_media1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 132650 | 19325018 | 14609115 | 65421 | 40483 | coronavirus | 28082.512980 | 77821.487020 | 222.812080 | -133.846480 | 6.859342e+06 | 2.707479e+07 | 4759.645684 | -2395.704704 |
| 85116 | 11838300 | 9946661 | 39230 | 36723 | belgique | 20140.420648 | 55812.579352 | 134.512328 | -80.803526 | 4.403545e+06 | 1.738142e+07 | 3542.953148 | -1783.298608 |
| 151790 | 3824512 | 613719 | 7715 | 2487 | diable | 2705.259456 | 7496.740544 | 96.318735 | -57.860075 | 8.971304e+05 | 3.541101e+06 | 3090.662217 | -1555.643950 |
| 84949 | 7968684 | 6626220 | 29100 | 25891 | belge | 14581.937144 | 40409.062856 | 120.226737 | -72.221962 | 2.950169e+06 | 1.164473e+07 | 2921.808490 | -1470.653660 |
| 223337 | 3107469 | 359102 | 3590 | 757 | hazard | 1152.691909 | 3194.308091 | 71.788323 | -43.124297 | 7.007220e+05 | 2.765849e+06 | 2875.130355 | -1447.158838 |
| 285657 | 2778591 | 498829 | 3319 | 1262 | madrid | 1214.741577 | 3366.258423 | 60.374998 | -36.268145 | 6.624876e+05 | 2.614932e+06 | 2599.848166 | -1308.599189 |
| 169632 | 2568809 | 355647 | 3001 | 1179 | eden | 1108.408599 | 3071.591401 | 56.846917 | -34.148776 | 5.911406e+05 | 2.333315e+06 | 2572.219887 | -1294.692860 |
| 376721 | 2870626 | 635009 | 3426 | 1682 | real | 1354.485915 | 3753.514085 | 56.286033 | -33.811845 | 7.086183e+05 | 2.797017e+06 | 2568.331287 | -1292.735585 |
| 389330 | 3887735 | 2284497 | 10170 | 10484 | rouge | 5476.811292 | 15177.188708 | 63.416771 | -38.095384 | 1.247636e+06 | 4.924596e+06 | 2363.612283 | -1189.692981 |
| 282524 | 1843184 | 306054 | 2250 | 812 | lukaku | 811.949074 | 2250.050926 | 50.467277 | -30.316432 | 4.344404e+05 | 1.714798e+06 | 2137.306780 | -1075.785100 |
| 270878 | 2243930 | 1023084 | 8705 | 5258 | league | 3702.562025 | 10260.437975 | 82.211116 | -49.385421 | 6.603842e+05 | 2.606630e+06 | 1948.643030 | -980.823696 |
| 445838 | 2212544 | 1126940 | 10607 | 4776 | trump | 4079.102746 | 11303.897254 | 102.209435 | -61.398705 | 6.750331e+05 | 2.664451e+06 | 1871.352869 | -941.920716 |
| 101606 | 1594791 | 421083 | 1952 | 413 | bruyne | 627.125918 | 1737.874082 | 52.905062 | -31.780846 | 4.074826e+05 | 1.608391e+06 | 1859.983144 | -936.197916 |
| 61895 | 4996615 | 6088391 | 21922 | 19685 | annoncer | 11032.908271 | 30574.091729 | 103.668461 | -62.275163 | 2.240689e+06 | 8.844317e+06 | 1841.097203 | -926.691927 |
| 60496 | 1647774 | 504751 | 8958 | 3632 | anderlecht | 3338.484273 | 9251.515727 | 97.257751 | -58.424156 | 4.351048e+05 | 1.717420e+06 | 1838.422810 | -925.345806 |
| 134360 | 1191936 | 102166 | 1455 | 352 | courtois | 479.161325 | 1327.838675 | 44.579701 | -26.779679 | 2.615858e+05 | 1.032516e+06 | 1819.027869 | -915.583619 |
| 134732 | 5833072 | 7996479 | 24467 | 29475 | covid-19 | 14303.774316 | 39638.225684 | 84.977949 | -51.047499 | 2.795463e+06 | 1.103409e+07 | 1816.791029 | -914.457736 |
| 90053 | 1594174 | 509468 | 6908 | 2791 | bilan | 2571.879187 | 7127.120813 | 85.501964 | -51.362282 | 4.252238e+05 | 1.678418e+06 | 1792.616006 | -902.289557 |
| 298826 | 5825958 | 8170334 | 22525 | 42177 | mesure | 17156.998365 | 47545.001635 | 40.981890 | -24.618422 | 2.829168e+06 | 1.116712e+07 | 1781.669025 | -896.779540 |
| 388272 | 1201758 | 170355 | 1448 | 339 | romelu | 473.857934 | 1313.142066 | 44.750534 | -26.882301 | 2.773547e+05 | 1.094758e+06 | 1755.269058 | -883.491465 |
| 356078 | 4406320 | 5342504 | 14062 | 17732 | police | 8430.799759 | 23363.200241 | 61.329099 | -36.841288 | 1.970597e+06 | 7.778227e+06 | 1735.117397 | -873.348393 |
| 163866 | 2163216 | 1355137 | 8557 | 5302 | décès | 3674.984395 | 10184.015605 | 80.532543 | -48.377078 | 7.111891e+05 | 2.807164e+06 | 1721.797668 | -866.644084 |
| 59979 | 13791596 | 29344381 | 39789 | 64040 | an | 27532.286223 | 76296.713777 | 73.867416 | -44.373239 | 8.719374e+06 | 3.441660e+07 | 1717.732637 | -864.598004 |
| 282842 | 4801225 | 6327805 | 21076 | 26267 | lundi | 12553.920645 | 34789.079355 | 76.059924 | -45.690310 | 2.249588e+06 | 8.879442e+06 | 1701.247053 | -856.300203 |
| 487072 | 2305015 | 1620938 | 8960 | 6472 | épidémie | 4092.096052 | 11339.903948 | 76.097272 | -45.712746 | 7.935801e+05 | 3.132373e+06 | 1696.656995 | -853.989859 |
| 103379 | 3235828 | 3218756 | 5938 | 8943 | but | 3945.987646 | 10935.012354 | 31.711309 | -19.049448 | 1.304710e+06 | 5.149874e+06 | 1690.642870 | -850.962728 |
| 298740 | 1432069 | 503902 | 1451 | 666 | messi | 561.363877 | 1555.636123 | 37.548291 | -22.555809 | 3.913312e+05 | 1.544640e+06 | 1663.677117 | -837.389873 |
| 101480 | 4841672 | 6706056 | 18777 | 31381 | bruxelles | 13300.372847 | 36857.627153 | 47.487714 | -28.526566 | 2.334222e+06 | 9.213506e+06 | 1641.198467 | -826.075542 |
| 232582 | 3917914 | 4788672 | 10125 | 10201 | hôpital | 5389.835689 | 14936.164311 | 64.498155 | -38.744986 | 1.759923e+06 | 6.946663e+06 | 1626.682951 | -818.769349 |
| 397846 | 4780848 | 6655026 | 17251 | 26179 | samedi | 11516.312308 | 31913.687692 | 53.438352 | -32.101201 | 2.311613e+06 | 9.124261e+06 | 1624.070887 | -817.454601 |
| 302676 | 4377445 | 5782810 | 17955 | 22984 | ministre | 10855.775031 | 30083.224969 | 68.136610 | -40.930660 | 2.053763e+06 | 8.106492e+06 | 1621.442604 | -816.131690 |
| 482163 | 970078 | 99946 | 795 | 261 | zidane | 280.019014 | 775.980986 | 30.774955 | -18.486967 | 2.162914e+05 | 8.537326e+05 | 1620.797757 | -815.807114 |
| 249970 | 3958967 | 4938957 | 15846 | 19397 | jeudi | 9345.369438 | 25897.630562 | 67.244574 | -40.394801 | 1.798599e+06 | 7.099325e+06 | 1610.870123 | -810.810171 |
| 132173 | 1729966 | 964487 | 391 | 507 | cordy | 238.122230 | 659.877770 | 9.907050 | -5.951310 | 5.446485e+05 | 2.149805e+06 | 1606.114700 | -808.416592 |
| 188310 | 4489604 | 6115559 | 14809 | 21250 | face | 9561.747767 | 26497.252233 | 53.661561 | -32.235286 | 2.143695e+06 | 8.461468e+06 | 1602.247543 | -806.470109 |
| 457748 | 4694112 | 6621349 | 18947 | 26733 | vendredi | 12112.943731 | 33567.056269 | 62.094581 | -37.301124 | 2.287273e+06 | 9.028188e+06 | 1591.432191 | -801.026345 |
| 157498 | 1427126 | 616647 | 6287 | 2621 | donald | 2362.130095 | 6545.869905 | 80.755732 | -48.511151 | 4.131220e+05 | 1.630651e+06 | 1577.613056 | -794.070666 |
| 163879 | 2690518 | 2578659 | 6311 | 4299 | décéder | 2813.448621 | 7796.551379 | 65.939342 | -39.610729 | 1.065095e+06 | 4.204082e+06 | 1574.969272 | -792.739952 |
| 89157 | 1111997 | 273703 | 3794 | 1663 | biden | 1447.030078 | 4009.969922 | 61.697667 | -37.062692 | 2.801012e+05 | 1.105599e+06 | 1571.851471 | -791.170649 |
| 325957 | 8847306 | 17145420 | 41255 | 65168 | nouveau | 28220.135961 | 78202.864039 | 77.593788 | -46.611725 | 5.254090e+06 | 2.073864e+07 | 1567.597509 | -789.029473 |
| 356138 | 2561931 | 2379873 | 5930 | 6846 | policier | 3387.805804 | 9388.194196 | 43.676665 | -26.237212 | 9.989211e+05 | 3.942883e+06 | 1563.853699 | -787.145076 |
| 61795 | 1740244 | 1082751 | 439 | 806 | annie | 330.136054 | 914.863946 | 5.991528 | -3.599199 | 5.706316e+05 | 2.252363e+06 | 1548.331971 | -779.332419 |
| 109634 | 4144422 | 5605739 | 19009 | 22028 | cas | 10881.761644 | 30155.238356 | 77.910034 | -46.801698 | 1.970868e+06 | 7.779293e+06 | 1548.252401 | -779.292369 |
| 298109 | 3672515 | 4664767 | 17463 | 19992 | mercredi | 9931.924419 | 27523.075581 | 75.568413 | -45.395053 | 1.685273e+06 | 6.652009e+06 | 1530.790458 | -770.503131 |
| 436217 | 859396 | 85572 | 1184 | 463 | thibaut | 436.734202 | 1210.265798 | 35.757447 | -21.480022 | 1.910129e+05 | 7.539551e+05 | 1529.304772 | -769.755330 |
| 59873 | 2115405 | 1717005 | 10378 | 8005 | américain | 4874.611309 | 13508.388691 | 78.824312 | -47.350919 | 7.746716e+05 | 3.057738e+06 | 1523.293532 | -766.729652 |
| 258479 | 1320595 | 570812 | 1999 | 1103 | kevin | 822.555855 | 2279.444145 | 41.019335 | -24.640915 | 3.823232e+05 | 1.509084e+06 | 1517.446254 | -763.786502 |
| 117186 | 1866221 | 1391811 | 7945 | 4551 | chiffre | 3313.558338 | 9182.441662 | 80.457936 | -48.332261 | 6.585686e+05 | 2.599463e+06 | 1488.131915 | -749.031517 |
| 419647 | 1562305 | 947363 | 9051 | 4036 | standard | 3470.273525 | 9616.726475 | 94.734656 | -56.908496 | 5.072966e+05 | 2.002371e+06 | 1481.238317 | -745.561716 |
| 366329 | 3230123 | 3949865 | 16141 | 20202 | président | 9637.055911 | 26705.944089 | 66.252857 | -39.799062 | 1.451341e+06 | 5.728647e+06 | 1476.515729 | -743.184664 |
fig = px.bar(graph_interactions_media1, x="res_inter_media", y="mot", labels={"res_inter_media":"Résiduel de Pearson (<b>en fonction des interactions</b>)","mot":"Lemme (dans les posts Facebook <b>médias</b>)"}, orientation="h", title="Lemmes les plus caractéristiques des posts Facebook <b>médias</b> {} en 2020<br>(en fonction des interactions suscitées)".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="gold", opacity=0.75, textfont_size=12)
fig
graph_interactions_nonmedia1 = khi2_1.sort_values(by=["res_inter_nonmedia"],ascending=False).head(50)
graph_interactions_nonmedia1
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | mot | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 358150 | 589111 | 19280588 | 2638 | 62212 | post | 17196.243454 | 47653.756546 | -111.017690 | 66.689952 | 4.016400e+06 | 1.585330e+07 | -1710.142326 | 860.777521 |
| 201650 | 450691 | 17394426 | 1252 | 53169 | from | 14430.790517 | 39990.209483 | -109.706029 | 65.902018 | 3.607157e+06 | 1.423796e+07 | -1661.952536 | 836.521828 |
| 94785 | 3274888 | 34801436 | 14457 | 68346 | bon | 21956.831869 | 60846.168131 | -50.613537 | 30.404293 | 7.696632e+06 | 3.037969e+07 | -1593.834420 | 802.235475 |
| 349213 | 3116610 | 31943611 | 12426 | 54765 | petit | 17817.005303 | 49373.994697 | -40.388005 | 24.261666 | 7.086966e+06 | 2.797326e+07 | -1491.417559 | 750.685303 |
| 340701 | 416861 | 14410454 | 2132 | 23092 | page | 6688.636004 | 18535.363996 | -55.715439 | 33.469080 | 2.997148e+06 | 1.183017e+07 | -1490.438034 | 750.192272 |
| 28573 | 89325 | 11357803 | 338 | 12178 | 3 | 3318.861728 | 9197.138272 | -51.742481 | 31.082466 | 2.313887e+06 | 9.133241e+06 | -1462.424593 | 736.092077 |
| 350552 | 2568016 | 28083848 | 8714 | 85736 | photo | 25045.261283 | 69404.738717 | -103.194593 | 61.990504 | 6.195874e+06 | 2.445599e+07 | -1457.467854 | 733.597168 |
| 59399 | 474552 | 12145865 | 1436 | 12786 | amour | 3771.240931 | 10450.759069 | -38.026780 | 22.843244 | 2.551052e+06 | 1.006936e+07 | -1300.086759 | 654.381475 |
| 488547 | 2070626 | 22454212 | 7169 | 51108 | être | 15453.284191 | 42823.715809 | -66.641474 | 40.032509 | 4.957376e+06 | 1.956746e+07 | -1296.531902 | 652.592185 |
| 188929 | 6160031 | 44825968 | 25377 | 102298 | faire | 33855.518627 | 93819.481373 | -46.079233 | 27.680470 | 1.030615e+07 | 4.067985e+07 | -1291.498314 | 650.058595 |
| 340959 | 359089 | 10552726 | 708 | 2920 | pairi | 962.035023 | 2665.964977 | -8.190269 | 4.920014 | 2.205681e+06 | 8.706134e+06 | -1243.367699 | 625.832687 |
| 480259 | 194471 | 9270260 | 662 | 6627 | youtube | 1932.820640 | 5356.179360 | -28.906032 | 17.364277 | 1.913172e+06 | 7.551559e+06 | -1242.576928 | 625.434663 |
| 433699 | 1921293 | 20189319 | 9523 | 41004 | temps | 13398.220401 | 37128.779599 | -33.479029 | 20.111343 | 4.469371e+06 | 1.764124e+07 | -1205.284479 | 606.664001 |
| 400404 | 1219960 | 15879803 | 6047 | 30383 | savoir | 9660.125660 | 26769.874340 | -36.761369 | 22.083093 | 3.456494e+06 | 1.364327e+07 | -1202.977669 | 605.502899 |
| 393678 | 8 | 7108753 | 1 | 680 | résa | 180.580444 | 500.419556 | -13.363607 | 8.027714 | 1.436943e+06 | 5.671818e+06 | -1198.718975 | 603.359342 |
| 152655 | 107371 | 7476766 | 386 | 10792 | dieu | 2964.064909 | 8213.935091 | -47.353272 | 28.445804 | 1.533034e+06 | 6.051103e+06 | -1151.439231 | 579.561708 |
| 457973 | 374426 | 9485232 | 2313 | 4866 | vent | 1903.651993 | 5275.348007 | 9.382075 | -5.635950 | 1.993001e+06 | 7.866657e+06 | -1146.513260 | 577.082286 |
| 53517 | 207929 | 8088620 | 622 | 8707 | aime | 2473.766464 | 6855.233536 | -37.231186 | 22.365319 | 1.677039e+06 | 6.619510e+06 | -1134.443145 | 571.006953 |
| 58920 | 1237266 | 14772671 | 2284 | 17228 | ami | 5173.987699 | 14338.012301 | -40.177536 | 24.135234 | 3.236200e+06 | 1.277374e+07 | -1111.170554 | 559.293002 |
| 53548 | 341488 | 8451007 | 1034 | 12027 | aimer | 3463.379117 | 9597.620883 | -41.280531 | 24.797820 | 1.777288e+06 | 7.015207e+06 | -1076.998373 | 542.092887 |
| 95118 | 59687 | 5794529 | 449 | 7101 | bonjour | 2002.029886 | 5547.970114 | -34.709194 | 20.850322 | 1.183353e+06 | 4.670863e+06 | -1032.952003 | 519.922729 |
| 241010 | 764384 | 10661017 | 1993 | 7370 | 2482.782228 | 6880.217772 | -9.829552 | 5.904756 | 2.309496e+06 | 9.115905e+06 | -1016.719792 | 511.752460 | |
| 233293 | 1293568 | 13928809 | 6202 | 33565 | ici | 10544.996352 | 29222.003648 | -42.292781 | 25.405893 | 3.077005e+06 | 1.214537e+07 | -1016.701852 | 511.743431 |
| 413622 | 299181 | 7447541 | 1776 | 6351 | soleil | 2155.032699 | 5971.967301 | -8.164879 | 4.904762 | 1.565899e+06 | 6.180823e+06 | -1012.273808 | 509.514633 |
| 435543 | 570475 | 9326705 | 2785 | 26021 | the | 7638.473229 | 21167.526771 | -55.532773 | 33.359350 | 2.000586e+06 | 7.896594e+06 | -1011.092916 | 508.920247 |
| 56512 | 5433 | 5089805 | 16 | 9989 | allah | 2653.021060 | 7351.978940 | -51.196851 | 30.754698 | 1.029936e+06 | 4.065302e+06 | -1009.504069 | 508.120522 |
| 441641 | 493 | 4985044 | 11 | 199 | touré | 55.685599 | 154.314401 | -5.988198 | 3.597198 | 1.007761e+06 | 3.977776e+06 | -1003.381993 | 505.039055 |
| 141836 | 364401 | 7792773 | 710 | 2192 | daiza | 769.521951 | 2132.478049 | -2.145688 | 1.288946 | 1.648866e+06 | 6.508308e+06 | -1000.298527 | 503.487033 |
| 84707 | 1328658 | 13810089 | 5437 | 25223 | bel | 8130.097522 | 22529.902478 | -29.867867 | 17.942065 | 3.060100e+06 | 1.207865e+07 | -989.783322 | 498.194344 |
| 158727 | 0 | 4731876 | 0 | 90 | dougourouni | 23.865257 | 66.134743 | -4.885208 | 2.934616 | 9.564869e+05 | 3.775389e+06 | -978.001495 | 492.264118 |
| 124635 | 93609 | 5331268 | 354 | 6979 | collection | 1944.488099 | 5388.511901 | -36.068475 | 21.666861 | 1.096568e+06 | 4.328309e+06 | -957.779201 | 482.085493 |
| 75786 | 1652840 | 15207630 | 5267 | 37873 | avoir | 11439.413147 | 31700.586853 | -57.710271 | 34.667405 | 3.408124e+06 | 1.345235e+07 | -950.801064 | 478.573140 |
| 276731 | 0 | 4456691 | 0 | 70 | likeetpartage | 18.561866 | 51.438134 | -4.308348 | 2.588088 | 9.008619e+05 | 3.555829e+06 | -949.137435 | 477.735775 |
| 461195 | 3212547 | 23340821 | 9320 | 42865 | vie | 13837.871467 | 38347.128533 | -38.406013 | 23.071055 | 5.367416e+06 | 2.118595e+07 | -930.118591 | 468.162891 |
| 118203 | 1009296 | 11210836 | 3449 | 17652 | chose | 5595.342068 | 15505.657932 | -28.693638 | 17.236689 | 2.470140e+06 | 9.749992e+06 | -929.486465 | 467.844719 |
| 343768 | 866330 | 10256637 | 3083 | 22249 | partager | 6717.274313 | 18614.725687 | -44.342601 | 26.637250 | 2.248362e+06 | 8.874605e+06 | -921.690449 | 463.920697 |
| 334523 | 5822 | 4162843 | 66 | 6442 | op | 1725.723244 | 4782.276756 | -39.953065 | 24.000391 | 8.426412e+05 | 3.326024e+06 | -911.612529 | 458.848110 |
| 403327 | 117352 | 5037946 | 997 | 3271 | sec | 1131.743517 | 3136.256483 | -4.005289 | 2.406036 | 1.042076e+06 | 4.113222e+06 | -905.862897 | 455.954108 |
| 343875 | 123193 | 4854726 | 839 | 6384 | partenaire | 1915.319452 | 5307.680548 | -24.593514 | 14.773684 | 1.006221e+06 | 3.971698e+06 | -880.294260 | 443.084473 |
| 259268 | 58739 | 4330704 | 242 | 1094 | kids | 354.266480 | 981.733520 | -5.964652 | 3.583054 | 8.872686e+05 | 3.502174e+06 | -879.590364 | 442.730176 |
| 61832 | 586371 | 8041785 | 1551 | 6874 | anniversaire | 2234.053217 | 6190.946783 | -14.451329 | 8.681125 | 1.744069e+06 | 6.884087e+06 | -876.624208 | 441.237201 |
| 347439 | 84880 | 4510160 | 276 | 4191 | pensée | 1184.512251 | 3282.487749 | -26.397391 | 15.857300 | 9.288273e+05 | 3.666213e+06 | -875.684871 | 440.764398 |
| 147261 | 525518 | 7544571 | 2582 | 14772 | demain | 4601.751872 | 12752.248128 | -29.773948 | 17.885646 | 1.631263e+06 | 6.438826e+06 | -865.751019 | 435.764325 |
| 333717 | 4701 | 3684615 | 46 | 1954 | ons | 530.339043 | 1469.660957 | -21.031618 | 12.634001 | 7.457470e+05 | 2.943569e+06 | -858.122761 | 431.924742 |
| 127843 | 234299 | 5490803 | 1398 | 11826 | concours | 3506.601749 | 9717.398251 | -35.608308 | 21.390432 | 1.157255e+06 | 4.567847e+06 | -857.958631 | 431.842130 |
| 237026 | 419335 | 6675426 | 1569 | 19993 | in | 5717.585217 | 15844.414783 | -54.864786 | 32.958081 | 1.434113e+06 | 5.660648e+06 | -847.382381 | 426.518714 |
| 201597 | 327951 | 6041988 | 1314 | 4784 | froid | 1617.003741 | 4480.996259 | -7.535160 | 4.526481 | 1.287600e+06 | 5.082339e+06 | -845.710886 | 425.677389 |
| 322758 | 1315 | 3541844 | 27 | 3088 | nl | 826.003059 | 2288.996941 | -27.800821 | 16.700360 | 7.162033e+05 | 2.826956e+06 | -844.734115 | 425.185744 |
| 113184 | 0 | 3524700 | 0 | 41 | cesttouss | 10.871950 | 30.128050 | -3.297264 | 1.980715 | 7.124721e+05 | 2.812228e+06 | -844.080603 | 424.856808 |
| 154006 | 561896 | 7570124 | 1847 | 11635 | dire | 3575.015486 | 9906.984514 | -28.900720 | 17.361086 | 1.643782e+06 | 6.488238e+06 | -843.838408 | 424.734902 |
fig = px.bar(graph_interactions_nonmedia1, x="res_inter_nonmedia", y="mot", labels={"res_inter_nonmedia":"Résiduel de Pearson (<b>en fonction des interactions</b>)","mot":"Lemme (dans les posts Facebook <b>non-médias</b>)"}, orientation="h", title="Lemmes les plus caractéristiques des posts Facebook <b>non-médias</b> {} en 2020<br>(en fonction des interactions suscitées)".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="lightyellow", opacity=0.75, textfont_size=12)
fig
media2 = pan.read_csv("belgique-bigrammes-media-nettoye.csv", low_memory=False, names=["bigramme","interactions"])
nonmedia2 = pan.read_csv("belgique-bigrammes-nonmedia-nettoye.csv", low_memory=False, names=["bigramme","interactions"])
media2
| bigramme | interactions | |
|---|---|---|
| 0 | vie bienvenue | 175299 |
| 1 | bienvenue émilie | 175299 |
| 2 | émilie dylan | 175299 |
| 3 | dylan faire | 175299 |
| 4 | faire tour | 175299 |
| ... | ... | ... |
| 8111423 | gardien timon | 7 |
| 8111424 | timon wellenreuther | 7 |
| 8111425 | wellenreuther club | 7 |
| 8111426 | club bruxellois | 7 |
| 8111427 | bruxellois renforcer | 7 |
8111428 rows × 2 columns
nonmedia2
| bigramme | interactions | |
|---|---|---|
| 0 | moment tendresse | 224233 |
| 1 | tendresse bébé | 224233 |
| 2 | bébé bébé | 224233 |
| 3 | bébé apprenon | 224233 |
| 4 | apprenon changer | 224233 |
| ... | ... | ... |
| 22718060 | voter ici | 5 |
| 22718061 | ici visitgaume | 5 |
| 22718062 | visitgaume soleildegaume | 5 |
| 22718063 | soleildegaume voter | 5 |
| 22718064 | voter logo | 5 |
22718065 rows × 2 columns
media2_table = pan.pivot_table(media2,index=["bigramme"],values=["bigramme","interactions"],aggfunc=[len,np.sum])
nonmedia2_table = pan.pivot_table(nonmedia2,index=["bigramme"],values=["bigramme","interactions"],aggfunc=[len,np.sum])
media2_table
| len | sum | |
|---|---|---|
| interactions | interactions | |
| bigramme | ||
| #balades travail | 1 | 45 |
| #breath4life masque | 1 | 9 |
| #cestbienlecas lidar | 1 | 422 |
| #confinemer coronavirus | 1 | 49 |
| #mobilité écologie | 1 | 26 |
| ... | ... | ... |
| 그ー ln24be | 1 | 34 |
| 그切 ln24be | 1 | 36 |
| 나4 oo | 1 | 5 |
| 무를 oand | 1 | 414 |
| 제이행 chadwick | 1 | 2645 |
2915288 rows × 2 columns
nonmedia2_table
| len | sum | |
|---|---|---|
| interactions | interactions | |
| bigramme | ||
| #1 cafes | 1 | 79 |
| #2 continue | 1 | 6 |
| #2 spectacle | 2 | 18 |
| #2020toussolidaire vivaforlife | 1 | 826 |
| #21juillet belgique | 1 | 693 |
| ... | ... | ... |
| single streaming | 1 | 12 |
| solidaire en | 1 | 12 |
| watch concert | 1 | 13 |
| 𐀁мвя иσιя | 1 | 12 |
| 𐌾ᗩïᗩᙅ lumière | 1 | 11 |
7862831 rows × 2 columns
media2_table["type"] = "media"
nonmedia2_table["type"] = "non-media"
media2_table
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| bigramme | |||
| #balades travail | 1 | 45 | media |
| #breath4life masque | 1 | 9 | media |
| #cestbienlecas lidar | 1 | 422 | media |
| #confinemer coronavirus | 1 | 49 | media |
| #mobilité écologie | 1 | 26 | media |
| ... | ... | ... | ... |
| 그ー ln24be | 1 | 34 | media |
| 그切 ln24be | 1 | 36 | media |
| 나4 oo | 1 | 5 | media |
| 무를 oand | 1 | 414 | media |
| 제이행 chadwick | 1 | 2645 | media |
2915288 rows × 3 columns
nonmedia2_table
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| bigramme | |||
| #1 cafes | 1 | 79 | non-media |
| #2 continue | 1 | 6 | non-media |
| #2 spectacle | 2 | 18 | non-media |
| #2020toussolidaire vivaforlife | 1 | 826 | non-media |
| #21juillet belgique | 1 | 693 | non-media |
| ... | ... | ... | ... |
| single streaming | 1 | 12 | non-media |
| solidaire en | 1 | 12 | non-media |
| watch concert | 1 | 13 | non-media |
| 𐀁мвя иσιя | 1 | 12 | non-media |
| 𐌾ᗩïᗩᙅ lumière | 1 | 11 | non-media |
7862831 rows × 3 columns
tableau2 = pan.concat([media2_table,nonmedia2_table], names = ["bigramme","nb","interactions","type"], levels=0)
tableau2
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| bigramme | |||
| #balades travail | 1 | 45 | media |
| #breath4life masque | 1 | 9 | media |
| #cestbienlecas lidar | 1 | 422 | media |
| #confinemer coronavirus | 1 | 49 | media |
| #mobilité écologie | 1 | 26 | media |
| ... | ... | ... | ... |
| single streaming | 1 | 12 | non-media |
| solidaire en | 1 | 12 | non-media |
| watch concert | 1 | 13 | non-media |
| 𐀁мвя иσιя | 1 | 12 | non-media |
| 𐌾ᗩïᗩᙅ lumière | 1 | 11 | non-media |
10778119 rows × 3 columns
tableau2["bigramme"] = tableau2.index
tableau2.columns = ["nb","interactions","media","bigramme"]
tableau2.reset_index(drop=True, inplace=True)
tableau2
| nb | interactions | media | bigramme | |
|---|---|---|---|---|
| 0 | 1 | 45 | media | #balades travail |
| 1 | 1 | 9 | media | #breath4life masque |
| 2 | 1 | 422 | media | #cestbienlecas lidar |
| 3 | 1 | 49 | media | #confinemer coronavirus |
| 4 | 1 | 26 | media | #mobilité écologie |
| ... | ... | ... | ... | ... |
| 10778114 | 1 | 12 | non-media | single streaming |
| 10778115 | 1 | 12 | non-media | solidaire en |
| 10778116 | 1 | 13 | non-media | watch concert |
| 10778117 | 1 | 12 | non-media | 𐀁мвя иσιя |
| 10778118 | 1 | 11 | non-media | 𐌾ᗩïᗩᙅ lumière |
10778119 rows × 4 columns
khi2_2 = pan.pivot_table(tableau2,index=["bigramme"],columns=["media"],values=["bigramme","interactions"],aggfunc=[len,np.sum],fill_value=0)
khi2_2
| len | sum | |||||||
|---|---|---|---|---|---|---|---|---|
| interactions | nb | interactions | nb | |||||
| media | media | non-media | media | non-media | media | non-media | media | non-media |
| bigramme | ||||||||
| #1 cafes | 0 | 1 | 0 | 1 | 0 | 79 | 0 | 1 |
| #2 continue | 0 | 1 | 0 | 1 | 0 | 6 | 0 | 1 |
| #2 spectacle | 0 | 1 | 0 | 1 | 0 | 18 | 0 | 2 |
| #2020toussolidaire vivaforlife | 0 | 1 | 0 | 1 | 0 | 826 | 0 | 1 |
| #21juillet belgique | 0 | 1 | 0 | 1 | 0 | 693 | 0 | 1 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... |
| single streaming | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| solidaire en | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| watch concert | 0 | 1 | 0 | 1 | 0 | 13 | 0 | 1 |
| 𐀁мвя иσιя | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| 𐌾ᗩïᗩᙅ lumière | 0 | 1 | 0 | 1 | 0 | 11 | 0 | 1 |
9562525 rows × 8 columns
khi2_2["bigramme"] = khi2_2.index
khi2_2.columns = khi2_2.columns.get_level_values(0)
khi2_2.reset_index(drop=True, inplace=True)
khi2_2.columns = ["a","b","c","d","interactions_media","interactions_nonmedia","obs_media","obs_nonmedia","bigramme"]
khi2_2 = khi2_2.drop(columns=["a","b","c","d"])
khi2_2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | |
|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes |
| 1 | 0 | 6 | 0 | 1 | #2 continue |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique |
| ... | ... | ... | ... | ... | ... |
| 9562520 | 0 | 12 | 0 | 1 | single streaming |
| 9562521 | 0 | 12 | 0 | 1 | solidaire en |
| 9562522 | 0 | 13 | 0 | 1 | watch concert |
| 9562523 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя |
| 9562524 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière |
9562525 rows × 5 columns
khi2_2.obs_media.sum(), khi2_2.obs_nonmedia.sum()
(8111428, 22718065)
print("Nb de lignes fichier media2 = ", media2.shape[0])
print("Somme observée média khi2_2 = ", khi2_2.obs_media.sum())
print("Nb de lignes fichier nonmedia2 = ", nonmedia2.shape[0])
print("Somme observée nonmédia khi2_2 = ", khi2_2.obs_nonmedia.sum())
Nb de lignes fichier media2 = 8111428 Somme observée média khi2_2 = 8111428 Nb de lignes fichier nonmedia2 = 22718065 Somme observée nonmédia khi2_2 = 22718065
khi2_2["exp_media"] = ((khi2_2.obs_media + khi2_2.obs_nonmedia) * khi2_2.obs_media.sum()) / (khi2_2.obs_media.sum() +khi2_2.obs_nonmedia.sum())
khi2_2["exp_nonmedia"] = ((khi2_2.obs_media + khi2_2.obs_nonmedia) * khi2_2.obs_nonmedia.sum()) / (khi2_2.obs_media.sum() +khi2_2.obs_nonmedia.sum())
khi2_2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | |
|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes | 0.263106 | 0.736894 |
| 1 | 0 | 6 | 0 | 1 | #2 continue | 0.263106 | 0.736894 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle | 0.526212 | 1.473788 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife | 0.263106 | 0.736894 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique | 0.263106 | 0.736894 |
| ... | ... | ... | ... | ... | ... | ... | ... |
| 9562520 | 0 | 12 | 0 | 1 | single streaming | 0.263106 | 0.736894 |
| 9562521 | 0 | 12 | 0 | 1 | solidaire en | 0.263106 | 0.736894 |
| 9562522 | 0 | 13 | 0 | 1 | watch concert | 0.263106 | 0.736894 |
| 9562523 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя | 0.263106 | 0.736894 |
| 9562524 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière | 0.263106 | 0.736894 |
9562525 rows × 7 columns
khi2_2["res_media"] = (khi2_2.obs_media - khi2_2.exp_media) / np.sqrt(khi2_2.exp_media)
khi2_2["res_nonmedia"] = (khi2_2.obs_nonmedia - khi2_2.exp_nonmedia) / np.sqrt(khi2_2.exp_nonmedia)
khi2_2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 1 | 0 | 6 | 0 | 1 | #2 continue | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle | 0.526212 | 1.473788 | -0.725405 | 0.433454 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 9562520 | 0 | 12 | 0 | 1 | single streaming | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 9562521 | 0 | 12 | 0 | 1 | solidaire en | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 9562522 | 0 | 13 | 0 | 1 | watch concert | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 9562523 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
| 9562524 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière | 0.263106 | 0.736894 | -0.512939 | 0.306499 |
9562525 rows × 9 columns
graph_media2 = khi2_2.sort_values(by=["res_media"],ascending=False).head(50)
graph_media2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 2324145 | 3052404 | 375839 | 7287 | 998 | coronavirus belgique | 2179.834128 | 6105.165872 | 109.387637 | -65.362887 |
| 2664379 | 2977596 | 151767 | 5464 | 1019 | diable rouge | 1705.716916 | 4777.283084 | 90.998872 | -54.374966 |
| 2801236 | 1343784 | 550471 | 6086 | 2338 | donald trump | 2216.405877 | 6207.594123 | 82.194123 | -49.113825 |
| 5986064 | 828623 | 89838 | 3868 | 1205 | nouveau cas | 1334.737300 | 3738.262700 | 69.339759 | -41.432899 |
| 5024298 | 194241 | 3907 | 2213 | 49 | libre afrique | 595.146023 | 1666.853977 | 66.317409 | -39.626941 |
| 6837077 | 480082 | 173768 | 3985 | 1790 | pro league | 1519.437790 | 4255.562210 | 63.252035 | -37.795274 |
| 3096480 | 2401438 | 236542 | 2583 | 485 | eden hazard | 807.209548 | 2260.790452 | 62.502670 | -37.347502 |
| 1528012 | 41529 | 2849 | 1930 | 34 | business am | 516.740402 | 1447.259598 | 62.170697 | -37.149137 |
| 9431033 | 999988 | 386521 | 3371 | 1585 | épidémie coronavirus | 1303.953885 | 3652.046115 | 57.242560 | -34.204405 |
| 234789 | 617764 | 341839 | 2669 | 931 | 24 heure | 947.181999 | 2652.818001 | 55.946191 | -33.429780 |
| 7226808 | 2255940 | 294881 | 2442 | 735 | real madrid | 835.888114 | 2341.111886 | 55.552259 | -33.194391 |
| 6269574 | 606184 | 309494 | 3205 | 1578 | pandémie coronavirus | 1258.436528 | 3524.563472 | 54.872268 | -32.788074 |
| 5056072 | 905840 | 607665 | 2876 | 1268 | ligue champion | 1090.311723 | 3053.688277 | 54.079206 | -32.314192 |
| 8606113 | 328448 | 58140 | 1847 | 333 | tour france | 573.571321 | 1606.428679 | 53.171745 | -31.771952 |
| 1663963 | 453261 | 89470 | 1997 | 460 | cas coronavirus | 646.451714 | 1810.548286 | 53.118044 | -31.739864 |
| 4755359 | 862091 | 193155 | 2644 | 1177 | joe biden | 1005.328449 | 2815.671551 | 51.681835 | -30.881680 |
| 8502943 | 561911 | 262722 | 2568 | 1150 | tester positif | 978.228520 | 2739.771480 | 50.829348 | -30.372290 |
| 8901993 | 190149 | 5485 | 1483 | 154 | vedia television | 430.704703 | 1206.295297 | 50.704654 | -30.297780 |
| 8445370 | 190149 | 5485 | 1483 | 154 | television local | 430.704703 | 1206.295297 | 50.704654 | -30.297780 |
| 5114411 | 190149 | 5485 | 1483 | 154 | local region | 430.704703 | 1206.295297 | 50.704654 | -30.297780 |
| 7288603 | 190149 | 5485 | 1483 | 154 | region verviers | 430.704703 | 1206.295297 | 50.704654 | -30.297780 |
| 8095073 | 959023 | 456198 | 2766 | 1384 | sophie wilmès | 1091.890360 | 3058.109640 | 50.663406 | -30.273133 |
| 480138 | 75535 | 14270 | 1449 | 143 | actualité région | 418.864928 | 1173.135072 | 50.333505 | -30.076006 |
| 7317733 | 464121 | 17642 | 1223 | 52 | remco evenepoel | 335.460291 | 939.539709 | 48.458195 | -28.955444 |
| 4873951 | 1251040 | 292038 | 1454 | 237 | kevin bruyne | 444.912433 | 1246.087567 | 47.840045 | -28.586078 |
| 1956022 | 330349 | 44153 | 1570 | 347 | club bruges | 504.374414 | 1412.625586 | 47.449115 | -28.352483 |
| 8047399 | 1239732 | 507756 | 2279 | 1053 | soin intensif | 876.669561 | 2455.330439 | 47.362258 | -28.300583 |
| 6716964 | 430545 | 162194 | 1850 | 608 | positif coronavirus | 646.714820 | 1811.285180 | 47.316451 | -28.273212 |
| 9372640 | 72109 | 13859 | 1286 | 134 | également bx1 | 373.610677 | 1046.389323 | 47.203104 | -28.205483 |
| 1538157 | 72074 | 13859 | 1284 | 134 | bx1be également | 373.084465 | 1044.915535 | 47.160079 | -28.179775 |
| 1538109 | 72074 | 13859 | 1284 | 134 | bx1 télévision | 373.084465 | 1044.915535 | 47.160079 | -28.179775 |
| 8769725 | 72074 | 13859 | 1284 | 134 | télévision bx1 | 373.084465 | 1044.915535 | 47.160079 | -28.179775 |
| 1628669 | 72074 | 13859 | 1284 | 134 | capitale bx1be | 373.084465 | 1044.915535 | 47.160079 | -28.179775 |
| 1538092 | 70936 | 14191 | 1274 | 138 | bx1 suivre | 371.505828 | 1040.494172 | 46.823254 | -27.978510 |
| 1538018 | 72074 | 14401 | 1284 | 153 | bx1 radio | 378.083481 | 1058.916519 | 46.590173 | -27.839236 |
| 7012119 | 412411 | 110784 | 1634 | 460 | président américain | 550.944196 | 1543.055804 | 46.142065 | -27.571476 |
| 8388268 | 1127253 | 1853299 | 5021 | 5103 | t il | 2663.686265 | 7460.313735 | 45.674716 | -27.292219 |
| 8281317 | 82914 | 49166 | 1365 | 248 | suivre actualité | 424.390157 | 1188.609843 | 45.659050 | -27.282858 |
| 7548601 | 325159 | 272895 | 2504 | 1437 | royaume uni | 1036.901183 | 2904.098817 | 45.560731 | -27.224109 |
| 5952698 | 208406 | 6033 | 1060 | 40 | nombre moyen | 289.416722 | 810.583278 | 45.295783 | -27.065793 |
| 7553890 | 385493 | 43255 | 1639 | 526 | rtc télé | 569.624730 | 1595.375270 | 44.805953 | -26.773103 |
| 3360543 | 560395 | 174700 | 1743 | 640 | europa league | 626.981862 | 1756.018138 | 44.570116 | -26.632182 |
| 1663998 | 277158 | 103426 | 1596 | 497 | cas covid-19 | 550.681090 | 1542.318910 | 44.544976 | -26.617160 |
| 5952762 | 235192 | 27081 | 1177 | 146 | nombre nouveau | 348.089385 | 974.910615 | 44.428568 | -26.547602 |
| 2260075 | 266380 | 30461 | 1203 | 167 | contamination coronavirus | 360.455372 | 1009.544628 | 44.377942 | -26.517352 |
| 8882029 | 460927 | 15646 | 1070 | 76 | van aert | 301.519603 | 844.480397 | 44.256291 | -26.444661 |
| 4551976 | 309311 | 15575 | 1035 | 56 | institut santé | 287.048767 | 803.951233 | 44.146415 | -26.379006 |
| 5988110 | 239701 | 23941 | 1168 | 158 | nouveau hospitalisation | 348.878703 | 977.121297 | 43.854179 | -26.204385 |
| 7515187 | 1168042 | 164697 | 1342 | 308 | romelu lukaku | 434.125083 | 1215.874917 | 43.573112 | -26.036438 |
| 5988286 | 220641 | 13853 | 1036 | 91 | nouveau infection | 296.520587 | 830.479413 | 42.943621 | -25.660295 |
fig = px.bar(graph_media2, x="res_media", y="bigramme", labels={"res_media":"Résiduel de Pearson","bigramme":"Bigramme (dans les posts Facebook <b>médias</b>)"}, orientation="h", title="Bigrammes les plus caractéristiques des posts Facebook <b>médias</b> {} en 2020".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="navy", opacity=0.75, textfont_size=12)
fig
graph_nonmedia2 = khi2_2.sort_values(by=["res_nonmedia"],ascending=False).head(50)
graph_nonmedia2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 6527313 | 406762 | 16320066 | 1094 | 48045 | photo from | 12928.771177 | 36210.228823 | -104.083345 | 62.193390 |
| 6786318 | 782096 | 5789921 | 408 | 11057 | prendre soin | 3016.511560 | 8448.488440 | -47.494166 | 28.379403 |
| 1649067 | 4976 | 495682 | 29 | 8099 | caroline taquin | 2138.526468 | 5989.473532 | -45.617099 | 27.257791 |
| 8547754 | 105277 | 2734947 | 740 | 11316 | timeline photo | 3172.007271 | 8883.992729 | -43.181506 | 25.802440 |
| 8414520 | 1064 | 275702 | 3 | 4518 | taquin bourgmestre | 1189.502727 | 3331.497273 | -34.402184 | 20.556492 |
| 5167404 | 0 | 80166 | 0 | 3878 | luc michel | 1020.325498 | 2857.674502 | -31.942534 | 19.086766 |
| 1425756 | 808 | 220043 | 6 | 3444 | bourgmestre caroline | 907.716082 | 2542.283918 | -29.929179 | 17.883717 |
| 6533982 | 39011 | 785119 | 98 | 3619 | photos from | 977.965414 | 2739.034586 | -28.138689 | 16.813837 |
| 2737145 | 38599 | 2282315 | 76 | 3416 | disneyland paris | 918.766539 | 2573.233461 | -27.803835 | 16.613751 |
| 8524534 | 38211 | 786173 | 1001 | 7841 | this is | 2326.384231 | 6515.615769 | -27.479010 | 16.419656 |
| 260284 | 269 | 638020 | 3 | 2769 | 3 3 | 729.330139 | 2042.669861 | -26.895027 | 16.070707 |
| 6086862 | 36162 | 742372 | 985 | 7536 | of post | 2241.927170 | 6279.072830 | -26.546017 | 15.862161 |
| 4653296 | 36162 | 742336 | 985 | 7535 | is re-share | 2241.664064 | 6278.335936 | -26.542017 | 15.859771 |
| 7225666 | 36162 | 742336 | 985 | 7535 | re-share of | 2241.664064 | 6278.335936 | -26.542017 | 15.859771 |
| 6071899 | 36335 | 703669 | 272 | 3929 | octobre 2020 | 1105.308771 | 3095.691229 | -25.064794 | 14.977079 |
| 7344888 | 64456 | 1217930 | 494 | 4963 | rendez vous | 1435.770046 | 4021.229954 | -24.854351 | 14.851332 |
| 1713380 | 22357 | 419968 | 452 | 4648 | centre culturel | 1341.841165 | 3758.158835 | -24.291925 | 14.515264 |
| 1950094 | 22479 | 1271855 | 54 | 2566 | cliquer lien | 689.338010 | 1930.661990 | -24.198516 | 14.459448 |
| 6783146 | 55399 | 458794 | 162 | 3162 | prendre contact | 874.564712 | 2449.435288 | -24.095078 | 14.397641 |
| 501905 | 30687 | 254990 | 253 | 3638 | administration communal | 1023.745877 | 2867.254123 | -24.088798 | 14.393888 |
| 5379768 | 140930 | 805252 | 342 | 3896 | mars 2020 | 1115.043697 | 3122.956303 | -23.150379 | 13.833150 |
| 7911006 | 36650 | 430894 | 258 | 3461 | septembre 2020 | 978.491626 | 2740.508374 | -23.032993 | 13.763008 |
| 8414526 | 895 | 115867 | 1 | 1955 | taquin caroline | 514.635553 | 1441.364447 | -22.641499 | 13.529077 |
| 5666925 | 1511 | 112827 | 3 | 1919 | moniteur automobile | 505.689945 | 1416.310055 | -22.354144 | 13.357372 |
| 2619536 | 4116 | 234044 | 19 | 2020 | der valk | 536.473360 | 1502.526640 | -22.341582 | 13.349866 |
| 2901048 | 19675 | 529568 | 228 | 3180 | décembre 2020 | 896.665625 | 2511.334375 | -22.330258 | 13.343100 |
| 3816823 | 0 | 138957 | 0 | 1894 | from commune | 498.322974 | 1395.677026 | -22.323149 | 13.338852 |
| 4423343 | 721 | 3529914 | 10 | 1918 | in het | 507.268582 | 1420.731418 | -22.078626 | 13.192741 |
| 5996771 | 31702 | 762862 | 224 | 3111 | novembre 2020 | 877.458879 | 2457.541121 | -22.059968 | 13.181592 |
| 2744853 | 7310 | 1106746 | 69 | 2272 | disponible ici | 615.931405 | 1725.068595 | -22.037721 | 13.168299 |
| 6249892 | 122256 | 1559209 | 500 | 4396 | page facebook | 1288.167518 | 3607.832482 | -21.960001 | 13.121859 |
| 4530134 | 1509 | 266275 | 29 | 1991 | inscrire vous | 531.474344 | 1488.525656 | -21.795796 | 13.023740 |
| 6249924 | 2344 | 242141 | 11 | 1765 | page fb | 467.276453 | 1308.723547 | -21.107709 | 12.612585 |
| 4318632 | 3564 | 175846 | 34 | 1908 | hésiter contacter | 510.952067 | 1431.047933 | -21.100107 | 12.608042 |
| 4490416 | 209 | 119769 | 9 | 1661 | info réservation | 439.387205 | 1230.612795 | -20.532208 | 12.268703 |
| 6087445 | 10203 | 506702 | 109 | 2197 | of the | 606.722691 | 1699.277309 | -20.206557 | 12.074115 |
| 3817947 | 0 | 201220 | 0 | 1506 | from ville | 396.237803 | 1109.762197 | -19.905723 | 11.894356 |
| 1380921 | 15633 | 3128883 | 90 | 2030 | bon journée | 557.784955 | 1562.215045 | -19.806733 | 11.835207 |
| 3862260 | 34398 | 249311 | 174 | 2428 | février 2020 | 684.602100 | 1917.397900 | -19.514772 | 11.660750 |
| 8883829 | 117 | 227365 | 3 | 1467 | van het | 386.765983 | 1083.234017 | -19.513822 | 11.660182 |
| 7821711 | 25111 | 615622 | 132 | 2201 | saviez vous | 613.826556 | 1719.173444 | -19.447684 | 11.620662 |
| 5504889 | 4826 | 317601 | 75 | 1872 | message priver | 512.267598 | 1434.732402 | -19.319632 | 11.544147 |
| 1081114 | 1111 | 128530 | 51 | 1730 | auvio voir | 468.591983 | 1312.408017 | -19.290999 | 11.527038 |
| 5747048 | 24 | 167366 | 1 | 1408 | mp page | 370.716510 | 1038.283490 | -19.202063 | 11.473895 |
| 3496426 | 30675 | 578356 | 179 | 2400 | faire plaisir | 678.550660 | 1900.449340 | -19.177341 | 11.459123 |
| 5075949 | 1297 | 76501 | 7 | 1438 | lire suite | 380.188330 | 1064.811670 | -19.139415 | 11.436461 |
| 6098902 | 7170 | 147127 | 77 | 1855 | offre emploi | 508.321006 | 1423.678994 | -19.130732 | 11.431272 |
| 546081 | 0 | 66070 | 0 | 1386 | afrique media | 364.665070 | 1021.334930 | -19.096206 | 11.410642 |
| 4799807 | 38279 | 553760 | 287 | 2887 | journée international | 835.098796 | 2338.901204 | -18.966619 | 11.333209 |
| 3282667 | 504 | 159211 | 7 | 1398 | envoyer mp | 369.664086 | 1035.335914 | -18.862572 | 11.271038 |
fig = px.bar(graph_nonmedia2, x="res_nonmedia", y="bigramme", labels={"res_nonmedia":"Résiduel de Pearson","bigramme":"Bigramme (dans les posts Facebook <b>non-médias</b>)"}, orientation="h", title="Bigrammes les plus caractéristiques des posts Facebook <b>non-médias</b> {} en 2020".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="aqua", opacity=0.75, textfont_size=12)
fig
khi2_2["exp_inter_media"] = ((khi2_2.interactions_media + khi2_2.interactions_nonmedia) * khi2_2.interactions_media.sum()) / (khi2_2.interactions_media.sum() + khi2_2.interactions_nonmedia.sum())
khi2_2["exp_inter_nonmedia"] = ((khi2_2.interactions_media + khi2_2.interactions_nonmedia) * khi2_2.interactions_nonmedia.sum()) / (khi2_2.interactions_media.sum() + khi2_2.interactions_nonmedia.sum())
khi2_2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 15.904769 | 63.095231 |
| 1 | 0 | 6 | 0 | 1 | #2 continue | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 1.207957 | 4.792043 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle | 0.526212 | 1.473788 | -0.725405 | 0.433454 | 3.623872 | 14.376128 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 166.295438 | 659.704562 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 139.519054 | 553.480946 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 9562520 | 0 | 12 | 0 | 1 | single streaming | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.415914 | 9.584086 |
| 9562521 | 0 | 12 | 0 | 1 | solidaire en | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.415914 | 9.584086 |
| 9562522 | 0 | 13 | 0 | 1 | watch concert | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.617241 | 10.382759 |
| 9562523 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.415914 | 9.584086 |
| 9562524 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.214588 | 8.785412 |
9562525 rows × 11 columns
khi2_2["res_inter_media"] = (khi2_2.interactions_media - khi2_2.exp_inter_media) / np.sqrt(khi2_2.exp_inter_media)
khi2_2["res_inter_nonmedia"] = (khi2_2.interactions_nonmedia - khi2_2.exp_inter_nonmedia) / np.sqrt(khi2_2.exp_inter_nonmedia)
khi2_2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 15.904769 | 63.095231 | -3.988078 | 2.002300 |
| 1 | 0 | 6 | 0 | 1 | #2 continue | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 1.207957 | 4.792043 | -1.099071 | 0.551812 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle | 0.526212 | 1.473788 | -0.725405 | 0.433454 | 3.623872 | 14.376128 | -1.903647 | 0.955767 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 166.295438 | 659.704562 | -12.895559 | 6.474490 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 139.519054 | 553.480946 | -11.811818 | 5.930375 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 9562520 | 0 | 12 | 0 | 1 | single streaming | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.415914 | 9.584086 | -1.554321 | 0.780380 |
| 9562521 | 0 | 12 | 0 | 1 | solidaire en | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.415914 | 9.584086 | -1.554321 | 0.780380 |
| 9562522 | 0 | 13 | 0 | 1 | watch concert | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.617241 | 10.382759 | -1.617789 | 0.812245 |
| 9562523 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.415914 | 9.584086 | -1.554321 | 0.780380 |
| 9562524 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière | 0.263106 | 0.736894 | -0.512939 | 0.306499 | 2.214588 | 8.785412 | -1.488149 | 0.747157 |
9562525 rows × 13 columns
graph_interactions_media2 = khi2_2.sort_values(by=["res_inter_media"],ascending=False).head(50)
graph_interactions_media2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2664379 | 2977596 | 151767 | 5464 | 1019 | diable rouge | 1705.716916 | 4777.283084 | 90.998872 | -54.374966 | 630022.749136 | 2.499340e+06 | 2957.610890 | -1484.931639 |
| 2324145 | 3052404 | 375839 | 7287 | 998 | coronavirus belgique | 2179.834128 | 6105.165872 | 109.387637 | -65.362887 | 690195.122639 | 2.738048e+06 | 2843.363305 | -1427.571202 |
| 3096480 | 2401438 | 236542 | 2583 | 485 | eden hazard | 807.209548 | 2260.790452 | 62.502670 | -37.347502 | 531094.478898 | 2.106886e+06 | 2566.465829 | -1288.548917 |
| 7226808 | 2255940 | 294881 | 2442 | 735 | real madrid | 835.888114 | 2341.111886 | 55.552259 | -33.194391 | 513547.088968 | 2.037274e+06 | 2431.397465 | -1220.734964 |
| 7515187 | 1168042 | 164697 | 1342 | 308 | romelu lukaku | 434.125083 | 1215.874917 | 43.573112 | -26.036438 | 268315.273320 | 1.064424e+06 | 1736.952428 | -872.074019 |
| 4873951 | 1251040 | 292038 | 1454 | 237 | kevin bruyne | 444.912433 | 1246.087567 | 47.840045 | -28.586078 | 310662.024090 | 1.232416e+06 | 1687.168139 | -847.078755 |
| 772852 | 1691969 | 944110 | 376 | 473 | annie cordy | 223.377088 | 625.622912 | 10.211756 | -6.101876 | 530711.757799 | 2.105367e+06 | 1594.039507 | -800.321538 |
| 2801236 | 1343784 | 550471 | 6086 | 2338 | donald trump | 2216.405877 | 6207.594123 | 82.194123 | -49.113825 | 381363.153672 | 1.512892e+06 | 1558.459986 | -782.458081 |
| 8521658 | 795022 | 53368 | 986 | 151 | thibaut courtois | 299.151648 | 837.848352 | 39.711397 | -23.728930 | 170803.131545 | 6.775869e+05 | 1510.389569 | -758.323303 |
| 3618277 | 715544 | 1710 | 7 | 4 | fils 15 | 2.894167 | 8.105833 | 2.413456 | -1.442123 | 144402.019487 | 5.728520e+05 | 1502.994702 | -754.610552 |
| 8047399 | 1239732 | 507756 | 2279 | 1053 | soin intensif | 876.669561 | 2455.330439 | 47.362258 | -28.300583 | 351815.111843 | 1.395673e+06 | 1496.976805 | -751.589138 |
| 5986064 | 828623 | 89838 | 3868 | 1205 | nouveau cas | 1334.737300 | 3738.262700 | 69.339759 | -41.432899 | 184910.259434 | 7.335507e+05 | 1496.964843 | -751.583133 |
| 5305313 | 893332 | 168442 | 1556 | 558 | manchester city | 556.206318 | 1557.793682 | 42.392832 | -25.331180 | 213762.920582 | 8.480111e+05 | 1469.831068 | -737.960042 |
| 4755359 | 862091 | 193155 | 2644 | 1177 | joe biden | 1005.328449 | 2815.671551 | 51.681835 | -30.881680 | 212448.663174 | 8.427973e+05 | 1409.442356 | -707.640601 |
| 3008445 | 652671 | 27257 | 36 | 40 | dénoncer voisin | 19.996064 | 56.003936 | 3.578941 | -2.138541 | 136887.317890 | 5.430407e+05 | 1394.074525 | -699.924854 |
| 5793883 | 715957 | 85819 | 28 | 51 | mère fils | 20.785383 | 58.214617 | 1.582466 | -0.945578 | 161418.512238 | 6.403575e+05 | 1380.241317 | -692.979596 |
| 9431033 | 999988 | 386521 | 3371 | 1585 | épidémie coronavirus | 1303.953885 | 3652.046115 | 57.242560 | -34.204405 | 279140.582886 | 1.107368e+06 | 1364.369041 | -685.010581 |
| 1686286 | 747304 | 161595 | 1256 | 671 | cause coronavirus | 507.005475 | 1419.994525 | 33.263835 | -19.876289 | 182985.178347 | 7.259138e+05 | 1319.218031 | -662.341554 |
| 5072005 | 701061 | 119255 | 796 | 159 | lionel messi | 251.266336 | 703.733664 | 34.365056 | -20.534307 | 165151.099914 | 6.551649e+05 | 1318.715299 | -662.089147 |
| 7497761 | 577739 | 45011 | 1207 | 317 | roberto martinez | 400.973713 | 1123.026287 | 40.252351 | -24.052169 | 125375.888647 | 4.973741e+05 | 1277.556660 | -641.424574 |
| 8095073 | 959023 | 456198 | 2766 | 1384 | sophie wilmès | 1091.890360 | 3058.109640 | 50.663406 | -30.273133 | 284921.060630 | 1.130300e+06 | 1262.883533 | -634.057617 |
| 3549067 | 637483 | 111473 | 955 | 309 | fc barcelone | 332.566124 | 931.433876 | 34.131409 | -20.394695 | 150784.462557 | 5.981715e+05 | 1253.377069 | -629.284695 |
| 2218210 | 1037905 | 583996 | 3962 | 4068 | conseil national | 2112.742069 | 5917.257931 | 40.232243 | -24.040154 | 326531.158848 | 1.295370e+06 | 1244.903180 | -625.030198 |
| 2033809 | 745120 | 244611 | 2429 | 1569 | comité concertation | 1051.898231 | 2946.101769 | 42.459920 | -25.371267 | 199258.777433 | 7.904722e+05 | 1222.850915 | -613.958388 |
| 5867131 | 1009985 | 576140 | 3854 | 3958 | national sécurité | 2055.384937 | 5756.615063 | 39.672689 | -23.705801 | 319328.512855 | 1.266796e+06 | 1222.202721 | -613.632948 |
| 90070 | 935660 | 527934 | 714 | 1208 | 15 an | 505.689945 | 1416.310055 | 9.263350 | -5.535171 | 294659.812716 | 1.168934e+06 | 1180.858071 | -592.874984 |
| 7317733 | 464121 | 17642 | 1223 | 52 | remco evenepoel | 335.460291 | 939.539709 | 48.458195 | -28.955444 | 96991.512232 | 3.847715e+05 | 1178.833321 | -591.858415 |
| 8882029 | 460927 | 15646 | 1070 | 76 | van aert | 301.519603 | 844.480397 | 44.256291 | -26.444661 | 95946.629274 | 3.806264e+05 | 1178.296661 | -591.588974 |
| 4311538 | 433990 | 2087 | 6 | 4 | héberger ami | 2.631061 | 7.368939 | 2.076958 | -1.241054 | 87793.723635 | 3.482833e+05 | 1168.397604 | -586.618941 |
| 4567993 | 515059 | 82980 | 702 | 230 | inter milan | 245.214895 | 686.785105 | 29.170140 | -17.430165 | 120400.917014 | 4.776381e+05 | 1137.381355 | -571.046572 |
| 4564018 | 408314 | 3920 | 27 | 7 | intensif cause | 8.945608 | 25.054392 | 6.036399 | -3.606957 | 82993.503140 | 3.292405e+05 | 1129.248126 | -566.963111 |
| 5989045 | 953952 | 638232 | 3479 | 3124 | nouveau mesure | 1737.289649 | 4865.710351 | 41.786868 | -24.969095 | 320548.348278 | 1.271636e+06 | 1118.751914 | -561.693263 |
| 5184949 | 550208 | 143010 | 1467 | 720 | lundi soir | 575.413064 | 1611.586936 | 37.168398 | -22.209400 | 139562.943037 | 5.536551e+05 | 1099.212200 | -551.882933 |
| 2852595 | 376861 | 148 | 30 | 5 | droit citer | 9.208714 | 25.791286 | 6.851440 | -4.093972 | 75901.787881 | 3.011072e+05 | 1092.399866 | -548.462656 |
| 7798373 | 819508 | 478945 | 3103 | 2711 | santé public | 1529.698928 | 4284.301072 | 40.226186 | -24.036534 | 261412.603359 | 1.037040e+06 | 1091.553884 | -548.037912 |
| 5056072 | 905840 | 607665 | 2876 | 1268 | ligue champion | 1090.311723 | 3053.688277 | 54.079206 | -32.314192 | 304708.204491 | 1.208797e+06 | 1088.999363 | -546.755360 |
| 9020867 | 440965 | 53711 | 1175 | 292 | vincent kompany | 385.976664 | 1081.023336 | 40.161444 | -23.997849 | 99591.237403 | 3.950848e+05 | 1081.731745 | -543.106498 |
| 726455 | 379175 | 8157 | 20 | 19 | an soin | 10.261138 | 28.738862 | 3.040258 | -1.816659 | 77980.078204 | 3.093519e+05 | 1078.588566 | -541.528398 |
| 3360543 | 560395 | 174700 | 1743 | 640 | europa league | 626.981862 | 1756.018138 | 44.570116 | -26.632182 | 147993.880153 | 5.871011e+05 | 1072.007830 | -538.224399 |
| 1493702 | 359272 | 73 | 4 | 2 | bruxelles contracter | 1.578637 | 4.421363 | 1.927165 | -1.151548 | 72345.561953 | 2.869994e+05 | 1066.754835 | -535.587021 |
| 836340 | 358602 | 356 | 5 | 4 | appel déchirant | 2.367955 | 6.632045 | 1.710435 | -1.022044 | 72267.648715 | 2.866904e+05 | 1065.127238 | -534.769851 |
| 5397630 | 357399 | 55 | 3 | 1 | masque plaire | 1.052424 | 2.947576 | 1.898449 | -1.134389 | 71964.854116 | 2.854891e+05 | 1064.009979 | -534.208907 |
| 8049029 | 356985 | 10 | 2 | 1 | soin témoignage | 0.789318 | 2.210682 | 1.362711 | -0.814267 | 71872.445392 | 2.851226e+05 | 1063.494212 | -533.949955 |
| 4322577 | 357098 | 92 | 6 | 3 | hôpital reine | 2.367955 | 6.632045 | 2.360285 | -1.410352 | 71911.704000 | 2.852783e+05 | 1063.478863 | -533.942249 |
| 3798531 | 356985 | 55 | 2 | 1 | françoise choquet | 0.789318 | 2.210682 | 1.362711 | -0.814267 | 71881.505070 | 2.851585e+05 | 1063.393399 | -533.899340 |
| 1852608 | 356985 | 55 | 2 | 1 | choquet mère | 0.789318 | 2.210682 | 1.362711 | -0.814267 | 71881.505070 | 2.851585e+05 | 1063.393399 | -533.899340 |
| 2910372 | 356985 | 55 | 2 | 1 | déchirant françoise | 0.789318 | 2.210682 | 1.362711 | -0.814267 | 71881.505070 | 2.851585e+05 | 1063.393399 | -533.899340 |
| 6593971 | 357321 | 803 | 3 | 2 | plaire appel | 1.315531 | 3.684469 | 1.468631 | -0.877558 | 72099.742667 | 2.860243e+05 | 1062.221366 | -533.310896 |
| 3463370 | 356985 | 587 | 2 | 8 | fabiola bruxelles | 2.631061 | 7.368939 | -0.389050 | 0.232471 | 71988.610607 | 2.855834e+05 | 1062.202850 | -533.301599 |
| 8923403 | 357624 | 2219 | 7 | 14 | venir transférer | 5.525228 | 15.474772 | 0.627407 | -0.374898 | 72445.822398 | 2.873972e+05 | 1059.521115 | -531.955177 |
fig = px.bar(graph_interactions_media2, x="res_inter_media", y="bigramme", labels={"res_inter_media":"Résiduel de Pearson (<b>en fonction des interactions</b>)","bigramme":"Bigramme (dans les posts Facebook <b>médias</b>)"}, orientation="h", title="Bigrammes les plus caractéristiques des posts Facebook <b>médias</b> {} en 2020<br>(en fonction des interactions suscitées)".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig
graph_interactions_nonmedia2 = khi2_2.sort_values(by=["res_inter_nonmedia"],ascending=False).head(50)
graph_interactions_nonmedia2
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | bigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 6527313 | 406762 | 16320066 | 1094 | 48045 | photo from | 12928.771177 | 36210.228823 | -104.083345 | 62.193390 | 3.367549e+06 | 1.335928e+07 | -1613.430175 | 810.057037 |
| 6256788 | 358840 | 7677739 | 700 | 2160 | pairi daiza | 752.483477 | 2107.516523 | -1.913261 | 1.143239 | 1.617974e+06 | 6.418605e+06 | -989.888222 | 496.994498 |
| 9083199 | 54407 | 4207833 | 182 | 597 | voice kids | 204.959660 | 574.040340 | -1.603730 | 0.958284 | 8.581006e+05 | 3.404139e+06 | -867.603704 | 435.598947 |
| 4423343 | 721 | 3529914 | 10 | 1918 | in het | 507.268582 | 1420.731418 | -22.078626 | 13.192741 | 7.108093e+05 | 2.819826e+06 | -842.239898 | 422.864507 |
| 5058006 | 0 | 3485133 | 0 | 39 | likeetpartage cesttouss | 10.261138 | 28.738862 | -3.203301 | 1.914083 | 7.016486e+05 | 2.783484e+06 | -837.644656 | 420.557368 |
| 6137739 | 115 | 3348925 | 5 | 880 | ook in | 232.848908 | 652.151092 | -14.931720 | 8.922218 | 6.742495e+05 | 2.674791e+06 | -820.986908 | 412.193991 |
| 9110551 | 115 | 3343761 | 5 | 838 | volg ons | 221.798451 | 621.201549 | -14.557169 | 8.698411 | 6.732098e+05 | 2.670666e+06 | -820.353494 | 411.875971 |
| 6134444 | 115 | 3343200 | 5 | 831 | ons ook | 219.956709 | 616.043291 | -14.493804 | 8.660548 | 6.730969e+05 | 2.670218e+06 | -820.284652 | 411.841408 |
| 8622968 | 0 | 3322292 | 0 | 36 | touré collection | 9.471820 | 26.528180 | -3.077632 | 1.838991 | 6.688644e+05 | 2.653428e+06 | -817.841312 | 410.614677 |
| 2540159 | 115 | 3321729 | 5 | 819 | daiza nl | 216.799435 | 607.200565 | -14.384532 | 8.595254 | 6.687742e+05 | 2.653070e+06 | -817.645545 | 410.516388 |
| 4198411 | 777 | 3323881 | 10 | 858 | het nederlands | 228.376104 | 639.623896 | -14.450397 | 8.634611 | 6.693407e+05 | 2.655317e+06 | -817.182752 | 410.284033 |
| 5886291 | 115 | 3290969 | 5 | 812 | nederlands op | 214.957692 | 602.042308 | -14.320405 | 8.556936 | 6.625814e+05 | 2.628503e+06 | -813.849767 | 408.610636 |
| 6139716 | 115 | 3228050 | 5 | 794 | op pairi | 210.221782 | 588.778218 | -14.154176 | 8.457609 | 6.499142e+05 | 2.578251e+06 | -806.029900 | 404.684506 |
| 8516822 | 182995 | 4632872 | 616 | 1461 | the voice | 546.471392 | 1530.528608 | 2.974266 | -1.777227 | 9.695602e+05 | 3.846307e+06 | -798.817063 | 401.063147 |
| 2006172 | 0 | 3121863 | 0 | 39 | collection likeetpartage | 10.261138 | 28.738862 | -3.203301 | 1.914083 | 6.285128e+05 | 2.493350e+06 | -792.787994 | 398.036124 |
| 6340536 | 0 | 3118599 | 0 | 34 | partenaire touré | 8.945608 | 25.054392 | -2.990921 | 1.787178 | 6.278557e+05 | 2.490743e+06 | -792.373443 | 397.827990 |
| 1380921 | 15633 | 3128883 | 90 | 2030 | bon journée | 557.784955 | 1562.215045 | -19.806733 | 11.835207 | 6.330734e+05 | 2.511443e+06 | -776.011265 | 389.613009 |
| 2540450 | 0 | 2870770 | 0 | 754 | daiza' post | 198.382008 | 555.617992 | -14.084815 | 8.416163 | 5.779612e+05 | 2.292809e+06 | -760.237597 | 381.693503 |
| 3817554 | 0 | 2870770 | 0 | 754 | from pairi | 198.382008 | 555.617992 | -14.084815 | 8.416163 | 5.779612e+05 | 2.292809e+06 | -760.237597 | 381.693503 |
| 6256789 | 0 | 2870770 | 0 | 754 | pairi daiza' | 198.382008 | 555.617992 | -14.084815 | 8.416163 | 5.779612e+05 | 2.292809e+06 | -760.237597 | 381.693503 |
| 2831025 | 0 | 2545312 | 0 | 31 | dougourouni tv | 8.156289 | 22.843711 | -2.855922 | 1.706512 | 5.124380e+05 | 2.032874e+06 | -715.847737 | 359.406627 |
| 5935146 | 0 | 2263822 | 0 | 617 | nl photo | 162.336470 | 454.663530 | -12.741133 | 7.613267 | 4.557667e+05 | 1.808055e+06 | -675.104934 | 338.950833 |
| 2831011 | 0 | 2065175 | 0 | 42 | dougourouni city | 11.050457 | 30.949543 | -3.324223 | 1.986338 | 4.157738e+05 | 1.649401e+06 | -644.805263 | 323.738237 |
| 4541686 | 0 | 2040068 | 0 | 29 | instagram partenaire | 7.630077 | 21.369923 | -2.762259 | 1.650545 | 4.107191e+05 | 1.629349e+06 | -640.873724 | 321.764324 |
| 2377882 | 5970 | 2048176 | 186 | 530 | cours après-midi | 188.383975 | 527.616025 | -0.173692 | 0.103787 | 4.135534e+05 | 1.640593e+06 | -633.797746 | 318.211678 |
| 2737145 | 38599 | 2282315 | 76 | 3416 | disneyland paris | 918.766539 | 2573.233461 | -27.803835 | 16.613751 | 4.672608e+05 | 1.853653e+06 | -627.097547 | 314.847701 |
| 551314 | 0 | 1925678 | 0 | 236 | age tendre | 62.093042 | 173.906958 | -7.879914 | 4.708520 | 3.876894e+05 | 1.537989e+06 | -622.647113 | 312.613265 |
| 1230400 | 27364 | 2136184 | 142 | 2020 | bel journée | 568.835411 | 1593.164589 | -17.896460 | 10.693752 | 4.355789e+05 | 1.727969e+06 | -618.522396 | 310.542363 |
| 8547754 | 105277 | 2734947 | 740 | 11316 | timeline photo | 3172.007271 | 8883.992729 | -43.181506 | 25.802440 | 5.718115e+05 | 2.268413e+06 | -616.960442 | 309.758151 |
| 5500321 | 0 | 1842355 | 0 | 252 | merveill monde | 66.302740 | 185.697260 | -8.142649 | 4.865514 | 3.709143e+05 | 1.471441e+06 | -609.027359 | 305.775177 |
| 4542421 | 325 | 1836805 | 4 | 136 | instagram youtube | 36.834856 | 103.165144 | -5.410104 | 3.232724 | 3.698624e+05 | 1.467268e+06 | -607.628736 | 305.072969 |
| 5833206 | 16442 | 1961324 | 311 | 478 | météo mons | 207.590721 | 581.409279 | 7.177208 | -4.288630 | 3.981761e+05 | 1.579590e+06 | -604.955411 | 303.730769 |
| 2169822 | 0 | 1768482 | 0 | 45 | confinement day | 11.839775 | 33.160225 | -3.440897 | 2.056055 | 3.560418e+05 | 1.412440e+06 | -596.692345 | 299.582120 |
| 3857503 | 2175 | 1747921 | 22 | 126 | félix radu | 38.939704 | 109.060296 | -2.714623 | 1.622081 | 3.523402e+05 | 1.397756e+06 | -589.918297 | 296.181065 |
| 1380473 | 42475 | 2085912 | 88 | 1645 | bon fête | 455.962890 | 1277.037110 | -17.232142 | 10.296800 | 4.285001e+05 | 1.699887e+06 | -589.712121 | 296.077550 |
| 8420461 | 0 | 1690743 | 0 | 50 | tarmac comedy | 13.155306 | 36.844694 | -3.627024 | 2.167272 | 3.403909e+05 | 1.350352e+06 | -583.430250 | 292.923603 |
| 8118663 | 28730 | 1912794 | 62 | 1141 | souhaite bon | 316.516651 | 886.483349 | -14.305990 | 8.548323 | 3.908796e+05 | 1.550644e+06 | -579.250656 | 290.825148 |
| 6446817 | 0 | 1652852 | 0 | 287 | pensée grégory | 75.511454 | 211.488546 | -8.689733 | 5.192415 | 3.327624e+05 | 1.320090e+06 | -576.855620 | 289.622669 |
| 1328466 | 0 | 1641157 | 0 | 292 | bisous 3 | 76.826984 | 215.173016 | -8.765100 | 5.237450 | 3.304079e+05 | 1.310749e+06 | -574.811183 | 288.596216 |
| 2671803 | 0 | 1614130 | 0 | 234 | dicton jour | 61.566830 | 172.433170 | -7.846453 | 4.688527 | 3.249667e+05 | 1.289163e+06 | -570.058465 | 286.210013 |
| 4070103 | 0 | 1613833 | 0 | 299 | grégory 3 | 78.668727 | 220.331273 | -8.869539 | 5.299856 | 3.249069e+05 | 1.288926e+06 | -570.006017 | 286.183680 |
| 4804353 | 40094 | 1942646 | 114 | 1316 | joyeux anniversaire | 376.241738 | 1053.758262 | -13.519740 | 8.078511 | 3.991775e+05 | 1.583562e+06 | -568.345498 | 285.349981 |
| 5245566 | 0 | 1559231 | 0 | 40 | mah touré | 10.524244 | 29.475756 | -3.244109 | 1.938467 | 3.139140e+05 | 1.245317e+06 | -560.280328 | 281.300691 |
| 683993 | 0 | 1544595 | 0 | 11 | america' got | 2.894167 | 8.105833 | -1.701225 | 1.016541 | 3.109674e+05 | 1.233628e+06 | -557.644543 | 279.977338 |
| 4831588 | 0 | 1542327 | 0 | 5 | jury america' | 1.315531 | 3.684469 | -1.146966 | 0.685352 | 3.105108e+05 | 1.231816e+06 | -557.234985 | 279.771711 |
| 1390057 | 0 | 1535706 | 0 | 384 | bonjour souhaite | 101.032747 | 282.967253 | -10.051505 | 6.006121 | 3.091778e+05 | 1.226528e+06 | -556.037631 | 279.170554 |
| 3992109 | 1067 | 1544509 | 5 | 25 | got talent | 7.893183 | 22.106817 | -1.029793 | 0.615337 | 3.111649e+05 | 1.234411e+06 | -555.908802 | 279.105872 |
| 5570953 | 3276 | 1536492 | 106 | 399 | min max | 132.868586 | 372.131414 | -2.330954 | 1.392825 | 3.099956e+05 | 1.229772e+06 | -550.888605 | 276.585375 |
| 2324982 | 176452 | 2826099 | 261 | 268 | coronavirus confinement | 139.183133 | 389.816867 | 10.325572 | -6.169886 | 6.044922e+05 | 2.398059e+06 | -550.540394 | 276.410549 |
| 6250714 | 0 | 1410106 | 0 | 165 | page merveill | 43.412508 | 121.587492 | -6.588817 | 3.937045 | 2.838913e+05 | 1.126215e+06 | -532.814487 | 267.510879 |
fig = px.bar(graph_interactions_nonmedia2, x="res_inter_nonmedia", y="bigramme", labels={"res_inter_nonmedia":"Résiduel de Pearson (<b>en fonction des interactions</b>)","bigramme":"Bigramme (dans les posts Facebook <b>non-médias</b>)"}, orientation="h", title="Bigrammes les plus caractéristiques des posts Facebook <b>non-médias</b> {} en 2020<br>(en fonction des interactions suscitées)".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="cyan", opacity=0.75, textfont_size=12)
fig
media3 = pan.read_csv("belgique-trigrammes-media-nettoye.csv", low_memory=False, names=["trigramme","interactions"])
nonmedia3 = pan.read_csv("belgique-trigrammes-nonmedia-nettoye.csv", low_memory=False, names=["trigramme","interactions"])
media3
| trigramme | interactions | |
|---|---|---|
| 0 | vie bienvenue émilie | 175299 |
| 1 | bienvenue émilie dylan | 175299 |
| 2 | émilie dylan faire | 175299 |
| 3 | dylan faire tour | 175299 |
| 4 | faire tour voiture | 175299 |
| ... | ... | ... |
| 7797166 | recrute gardien timon | 7 |
| 7797167 | gardien timon wellenreuther | 7 |
| 7797168 | timon wellenreuther club | 7 |
| 7797169 | wellenreuther club bruxellois | 7 |
| 7797170 | club bruxellois renforcer | 7 |
7797171 rows × 2 columns
nonmedia3
| trigramme | interactions | |
|---|---|---|
| 0 | moment tendresse bébé | 224233 |
| 1 | tendresse bébé bébé | 224233 |
| 2 | bébé bébé apprenon | 224233 |
| 3 | bébé apprenon changer | 224233 |
| 4 | apprenon changer regard | 224233 |
| ... | ... | ... |
| 22091456 | web voter ici | 5 |
| 22091457 | voter ici visitgaume | 5 |
| 22091458 | ici visitgaume soleildegaume | 5 |
| 22091459 | visitgaume soleildegaume voter | 5 |
| 22091460 | soleildegaume voter logo | 5 |
22091461 rows × 2 columns
media3_table = pan.pivot_table(media3,index=["trigramme"],values=["trigramme","interactions"],aggfunc=[len,np.sum])
nonmedia3_table = pan.pivot_table(nonmedia3,index=["trigramme"],values=["trigramme","interactions"],aggfunc=[len,np.sum])
media3_table
| len | sum | |
|---|---|---|
| interactions | interactions | |
| trigramme | ||
| #balades travail ecaussinne | 1 | 45 |
| #breath4life masque plongée | 1 | 9 |
| #cestbienlecas lidar enregistrer | 1 | 422 |
| #confinemer coronavirus coronavirus | 1 | 49 |
| #mobilité écologie train | 1 | 26 |
| ... | ... | ... |
| 静岡県 第3戦 8月22日・23 | 1 | 45 |
| 그切 ln24be françois | 1 | 36 |
| 나4 oo 6ဝ | 1 | 5 |
| 무를 oand egg | 1 | 414 |
| 제이행 chadwick boseman | 1 | 2645 |
5197497 rows × 2 columns
nonmedia3_table
| len | sum | |
|---|---|---|
| interactions | interactions | |
| trigramme | ||
| #1 cafes colombe | 1 | 79 |
| #2 continue faire | 1 | 6 |
| #2 spectacle danse | 2 | 18 |
| #2020toussolidaire vivaforlife vivaforreyer | 1 | 826 |
| #21juillet belgique belgië | 1 | 693 |
| ... | ... | ... |
| single musique eden | 1 | 73 |
| single streaming timeline | 1 | 12 |
| solidaire en 2021 | 1 | 12 |
| 𐀁мвя иσιя parfum | 1 | 12 |
| 𐌾ᗩïᗩᙅ lumière temps | 1 | 11 |
14650061 rows × 2 columns
media3_table["type"] = "media"
nonmedia3_table["type"] = "non-media"
media3_table
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| trigramme | |||
| #balades travail ecaussinne | 1 | 45 | media |
| #breath4life masque plongée | 1 | 9 | media |
| #cestbienlecas lidar enregistrer | 1 | 422 | media |
| #confinemer coronavirus coronavirus | 1 | 49 | media |
| #mobilité écologie train | 1 | 26 | media |
| ... | ... | ... | ... |
| 静岡県 第3戦 8月22日・23 | 1 | 45 | media |
| 그切 ln24be françois | 1 | 36 | media |
| 나4 oo 6ဝ | 1 | 5 | media |
| 무를 oand egg | 1 | 414 | media |
| 제이행 chadwick boseman | 1 | 2645 | media |
5197497 rows × 3 columns
nonmedia3_table
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| trigramme | |||
| #1 cafes colombe | 1 | 79 | non-media |
| #2 continue faire | 1 | 6 | non-media |
| #2 spectacle danse | 2 | 18 | non-media |
| #2020toussolidaire vivaforlife vivaforreyer | 1 | 826 | non-media |
| #21juillet belgique belgië | 1 | 693 | non-media |
| ... | ... | ... | ... |
| single musique eden | 1 | 73 | non-media |
| single streaming timeline | 1 | 12 | non-media |
| solidaire en 2021 | 1 | 12 | non-media |
| 𐀁мвя иσιя parfum | 1 | 12 | non-media |
| 𐌾ᗩïᗩᙅ lumière temps | 1 | 11 | non-media |
14650061 rows × 3 columns
tableau3 = pan.concat([media3_table,nonmedia3_table], names = ["trigramme","nb","interactions","type"], levels=0)
tableau3
| len | sum | type | |
|---|---|---|---|
| interactions | interactions | ||
| trigramme | |||
| #balades travail ecaussinne | 1 | 45 | media |
| #breath4life masque plongée | 1 | 9 | media |
| #cestbienlecas lidar enregistrer | 1 | 422 | media |
| #confinemer coronavirus coronavirus | 1 | 49 | media |
| #mobilité écologie train | 1 | 26 | media |
| ... | ... | ... | ... |
| single musique eden | 1 | 73 | non-media |
| single streaming timeline | 1 | 12 | non-media |
| solidaire en 2021 | 1 | 12 | non-media |
| 𐀁мвя иσιя parfum | 1 | 12 | non-media |
| 𐌾ᗩïᗩᙅ lumière temps | 1 | 11 | non-media |
19847558 rows × 3 columns
tableau3["trigramme"] = tableau3.index
tableau3.columns = ["nb","interactions","media","trigramme"]
tableau3.reset_index(drop=True, inplace=True)
tableau3
| nb | interactions | media | trigramme | |
|---|---|---|---|---|
| 0 | 1 | 45 | media | #balades travail ecaussinne |
| 1 | 1 | 9 | media | #breath4life masque plongée |
| 2 | 1 | 422 | media | #cestbienlecas lidar enregistrer |
| 3 | 1 | 49 | media | #confinemer coronavirus coronavirus |
| 4 | 1 | 26 | media | #mobilité écologie train |
| ... | ... | ... | ... | ... |
| 19847553 | 1 | 73 | non-media | single musique eden |
| 19847554 | 1 | 12 | non-media | single streaming timeline |
| 19847555 | 1 | 12 | non-media | solidaire en 2021 |
| 19847556 | 1 | 12 | non-media | 𐀁мвя иσιя parfum |
| 19847557 | 1 | 11 | non-media | 𐌾ᗩïᗩᙅ lumière temps |
19847558 rows × 4 columns
khi2_3 = pan.pivot_table(tableau3,index=["trigramme"],columns=["media"],values=["trigramme","interactions"],aggfunc=[len,np.sum],fill_value=0)
khi2_3
| len | sum | |||||||
|---|---|---|---|---|---|---|---|---|
| interactions | nb | interactions | nb | |||||
| media | media | non-media | media | non-media | media | non-media | media | non-media |
| trigramme | ||||||||
| #1 cafes colombe | 0 | 1 | 0 | 1 | 0 | 79 | 0 | 1 |
| #2 continue faire | 0 | 1 | 0 | 1 | 0 | 6 | 0 | 1 |
| #2 spectacle danse | 0 | 1 | 0 | 1 | 0 | 18 | 0 | 2 |
| #2020toussolidaire vivaforlife vivaforreyer | 0 | 1 | 0 | 1 | 0 | 826 | 0 | 1 |
| #21juillet belgique belgië | 0 | 1 | 0 | 1 | 0 | 693 | 0 | 1 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... |
| single musique eden | 0 | 1 | 0 | 1 | 0 | 73 | 0 | 1 |
| single streaming timeline | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| solidaire en 2021 | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| 𐀁мвя иσιя parfum | 0 | 1 | 0 | 1 | 0 | 12 | 0 | 1 |
| 𐌾ᗩïᗩᙅ lumière temps | 0 | 1 | 0 | 1 | 0 | 11 | 0 | 1 |
18981336 rows × 8 columns
khi2_3["trigramme"] = khi2_3.index
khi2_3.columns = khi2_3.columns.get_level_values(0)
khi2_3.reset_index(drop=True, inplace=True)
khi2_3.columns = ["a","b","c","d","interactions_media","interactions_nonmedia","obs_media","obs_nonmedia","trigramme"]
khi2_3 = khi2_3.drop(columns=["a","b","c","d"])
khi2_3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | |
|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes colombe |
| 1 | 0 | 6 | 0 | 1 | #2 continue faire |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle danse |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife vivaforreyer |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique belgië |
| ... | ... | ... | ... | ... | ... |
| 18981331 | 0 | 73 | 0 | 1 | single musique eden |
| 18981332 | 0 | 12 | 0 | 1 | single streaming timeline |
| 18981333 | 0 | 12 | 0 | 1 | solidaire en 2021 |
| 18981334 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя parfum |
| 18981335 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière temps |
18981336 rows × 5 columns
khi2_3.obs_media.sum(), khi2_3.obs_nonmedia.sum()
(7797171, 22091461)
print("Nb de lignes fichier media3 = ", media3.shape[0])
print("Somme observée média khi2_3 = ", khi2_3.obs_media.sum())
print("Nb de lignes fichier nonmedia3 = ", nonmedia3.shape[0])
print("Somme observée nonmédia khi2_3 = ", khi2_3.obs_nonmedia.sum())
Nb de lignes fichier media3 = 7797171 Somme observée média khi2_3 = 7797171 Nb de lignes fichier nonmedia3 = 22091461 Somme observée nonmédia khi2_3 = 22091461
khi2_3["exp_media"] = ((khi2_3.obs_media + khi2_3.obs_nonmedia) * khi2_3.obs_media.sum()) / (khi2_3.obs_media.sum() + khi2_3.obs_nonmedia.sum())
khi2_3["exp_nonmedia"] = ((khi2_3.obs_media + khi2_3.obs_nonmedia) * khi2_3.obs_nonmedia.sum()) / (khi2_3.obs_media.sum() + khi2_3.obs_nonmedia.sum())
khi2_3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | |
|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes colombe | 0.260874 | 0.739126 |
| 1 | 0 | 6 | 0 | 1 | #2 continue faire | 0.260874 | 0.739126 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle danse | 0.521748 | 1.478252 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife vivaforreyer | 0.260874 | 0.739126 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique belgië | 0.260874 | 0.739126 |
| ... | ... | ... | ... | ... | ... | ... | ... |
| 18981331 | 0 | 73 | 0 | 1 | single musique eden | 0.260874 | 0.739126 |
| 18981332 | 0 | 12 | 0 | 1 | single streaming timeline | 0.260874 | 0.739126 |
| 18981333 | 0 | 12 | 0 | 1 | solidaire en 2021 | 0.260874 | 0.739126 |
| 18981334 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя parfum | 0.260874 | 0.739126 |
| 18981335 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière temps | 0.260874 | 0.739126 |
18981336 rows × 7 columns
khi2_3["res_media"] = (khi2_3.obs_media - khi2_3.exp_media) / np.sqrt(khi2_3.exp_media)
khi2_3["res_nonmedia"] = (khi2_3.obs_nonmedia - khi2_3.exp_nonmedia) / np.sqrt(khi2_3.exp_nonmedia)
khi2_3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes colombe | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 1 | 0 | 6 | 0 | 1 | #2 continue faire | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle danse | 0.521748 | 1.478252 | -0.722321 | 0.429128 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife vivaforreyer | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique belgië | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 18981331 | 0 | 73 | 0 | 1 | single musique eden | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 18981332 | 0 | 12 | 0 | 1 | single streaming timeline | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 18981333 | 0 | 12 | 0 | 1 | solidaire en 2021 | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 18981334 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя parfum | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
| 18981335 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière temps | 0.260874 | 0.739126 | -0.510758 | 0.303439 |
18981336 rows × 9 columns
graph_media3 = khi2_3.sort_values(by=["res_media"],ascending=False).head(50)
graph_media3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 16768707 | 190149 | 5485 | 1483 | 154 | television local region | 427.050958 | 1209.949042 | 51.097907 | -30.357040 |
| 17639945 | 190149 | 5485 | 1483 | 154 | vedia television local | 427.050958 | 1209.949042 | 51.097907 | -30.357040 |
| 9994421 | 190149 | 5485 | 1483 | 154 | local region verviers | 427.050958 | 1209.949042 | 51.097907 | -30.357040 |
| 919457 | 72074 | 13859 | 1284 | 134 | actualité région bruxelles | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 2898345 | 72074 | 13859 | 1284 | 134 | bruxelles capitale bx1be | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 16477661 | 72074 | 13859 | 1284 | 134 | suivre actualité région | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 17401137 | 72074 | 13859 | 1284 | 134 | télévision bx1 radio | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 2990877 | 72074 | 13859 | 1284 | 134 | bx1 télévision bx1 | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 3146471 | 72074 | 13859 | 1284 | 134 | capitale bx1be également | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 2990991 | 72074 | 13859 | 1284 | 134 | bx1be également bx1 | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 18609877 | 72074 | 13859 | 1284 | 134 | également bx1 télévision | 369.919523 | 1048.080477 | 47.525950 | -28.234956 |
| 2990848 | 70905 | 13859 | 1272 | 134 | bx1 suivre actualité | 366.789033 | 1039.210967 | 47.265215 | -28.080054 |
| 15021172 | 321975 | 42982 | 1533 | 523 | rtc télé liège | 536.357220 | 1519.642780 | 43.034074 | -25.566352 |
| 5122140 | 254875 | 8005 | 1116 | 160 | dernier 24 heure | 332.875395 | 943.124605 | 42.922995 | -25.500361 |
| 18448921 | 409367 | 331747 | 1195 | 224 | yves van laethem | 370.180397 | 1048.819603 | 42.869880 | -25.468805 |
| 8887323 | 280842 | 14583 | 924 | 40 | institut santé public | 251.482666 | 712.517334 | 42.408163 | -25.194501 |
| 16678563 | 210812 | 13561 | 1014 | 121 | t on apprendre | 296.092142 | 838.907858 | 41.721053 | -24.786293 |
| 10421499 | 383013 | 58659 | 1149 | 236 | marc van ranst | 361.310676 | 1023.689324 | 41.439505 | -24.619026 |
| 4321599 | 1003102 | 559225 | 3831 | 3824 | conseil national sécurité | 1996.991498 | 5658.008502 | 41.040556 | -24.382012 |
| 18399250 | 334820 | 15156 | 789 | 55 | wout van aert | 220.177769 | 623.822231 | 38.334502 | -22.774358 |
| 15515939 | 160911 | 12762 | 749 | 38 | santé public sciensano | 205.307944 | 581.692056 | 37.944612 | -22.542725 |
| 9035126 | 103459 | 5176 | 767 | 67 | invité martin buxant | 217.569028 | 616.430972 | 37.248999 | -22.129465 |
| 11651603 | 131552 | 232 | 673 | 11 | nombre moyen infection | 178.437908 | 505.562092 | 37.023482 | -21.995487 |
| 15179723 | 85014 | 51095 | 1362 | 641 | région bruxelles capitale | 522.530891 | 1480.469109 | 36.723895 | -21.817503 |
| 1354814 | 160618 | 23130 | 829 | 152 | américain donald trump | 255.917526 | 725.082474 | 35.823426 | -21.282538 |
| 16906741 | 253368 | 56535 | 1031 | 337 | tester positif coronavirus | 356.875816 | 1011.124184 | 35.684642 | -21.200088 |
| 18720987 | 237912 | 4790 | 683 | 50 | épidémie coronavirus belgique | 191.220740 | 541.779260 | 35.563354 | -21.128031 |
| 11651734 | 116313 | 2611 | 650 | 45 | nombre nouveau cas | 181.307523 | 513.692477 | 34.808080 | -20.679326 |
| 7754060 | 309852 | 383296 | 1604 | 1076 | georges louis bouchez | 699.142680 | 1980.857320 | 34.221355 | -20.330755 |
| 13955128 | 143579 | 22518 | 734 | 123 | président américain donald | 223.569133 | 633.430867 | 34.137445 | -20.280904 |
| 16278853 | 136644 | 946 | 500 | 6 | sport foot magazine | 132.002312 | 373.997688 | 32.029807 | -19.028766 |
| 2575168 | 23623 | 0 | 473 | 0 | bilan jour belgique | 123.393466 | 349.606534 | 31.472661 | -18.697768 |
| 18878176 | 54812 | 4435 | 503 | 21 | évolution pandémie covid-19 | 136.698046 | 387.301954 | 31.329833 | -18.612915 |
| 11716317 | 105715 | 5090 | 541 | 56 | nouveau cas coronavirus | 155.741858 | 441.258142 | 30.870890 | -18.340259 |
| 7538015 | 312212 | 0 | 433 | 0 | from rtl info' | 112.958500 | 320.041500 | 30.112502 | -17.889704 |
| 15022690 | 312212 | 0 | 433 | 0 | rtl info' post | 112.958500 | 320.041500 | 30.112502 | -17.889704 |
| 5142326 | 49865 | 132 | 428 | 7 | dernier évolution pandémie | 113.480248 | 321.519752 | 29.524856 | -17.540586 |
| 8887065 | 121244 | 3086 | 511 | 65 | institut royal météorologique | 150.263501 | 425.736499 | 29.428175 | -17.483148 |
| 16906747 | 148224 | 84901 | 824 | 361 | tester positif covid-19 | 309.135849 | 875.864151 | 29.283162 | -17.396997 |
| 10962486 | 283986 | 136586 | 944 | 501 | ministre sophie wilmès | 376.963124 | 1068.036876 | 29.205313 | -17.350747 |
| 11155652 | 100218 | 20318 | 577 | 127 | mondial santé om | 183.655391 | 520.344609 | 29.024942 | -17.243589 |
| 5040933 | 139402 | 8394 | 448 | 29 | deceuninck quick step | 124.436962 | 352.563038 | 29.005757 | -17.232192 |
| 12123931 | 141289 | 102433 | 826 | 375 | organisation mondial santé | 313.309835 | 887.690165 | 28.964629 | -17.207758 |
| 4791124 | 76781 | 1444 | 418 | 23 | crise spf santé | 115.045493 | 325.954507 | 28.245067 | -16.780269 |
| 12885351 | 279752 | 7859 | 380 | 13 | photo from rtl | 102.523535 | 290.476465 | 27.404026 | -16.280610 |
| 11716323 | 71481 | 6723 | 465 | 74 | nouveau cas covid-19 | 140.611158 | 398.388842 | 27.356215 | -16.252206 |
| 14325887 | 67468 | 661 | 376 | 13 | rapport semaine précédent | 101.480038 | 287.519962 | 27.251074 | -16.189742 |
| 17607267 | 187714 | 1076 | 422 | 44 | van der poel | 121.567347 | 344.432653 | 27.248253 | -16.188066 |
| 9336973 | 21009 | 0 | 348 | 0 | jour belgique coronavirus | 90.784199 | 257.215801 | 26.995571 | -16.037949 |
| 7181215 | 154616 | 60364 | 526 | 146 | finale ligue champion | 175.307418 | 496.692582 | 26.486613 | -15.735580 |
fig = px.bar(graph_media3, x="res_media", y="trigramme", labels={"res_media":"Résiduel de Pearson","trigramme":"Trigramme (dans les posts Facebook <b>médias</b>)"}, orientation="h", title="Trigrammes les plus caractéristiques des posts Facebook <b>médias</b> {} en 2020".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="DarkOliveGreen", opacity=0.75, textfont_size=12)
fig
graph_nonmedia3 = khi2_3.sort_values(by=["res_nonmedia"],ascending=False).head(50)
graph_nonmedia3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|
| 3178379 | 1064 | 273967 | 3 | 4481 | caroline taquin bourgmestre | 1169.759618 | 3314.240382 | -34.114034 | 20.266996 |
| 2786711 | 808 | 219791 | 6 | 3436 | bourgmestre caroline taquin | 897.928770 | 2544.071230 | -29.765229 | 17.683391 |
| 16718669 | 0 | 201369 | 0 | 3050 | taquin bourgmestre caroline | 795.666110 | 2254.333890 | -28.207554 | 16.757983 |
| 9061281 | 36162 | 742336 | 985 | 7535 | is re-share of | 2222.647625 | 6297.352375 | -26.251958 | 15.596172 |
| 14370650 | 36162 | 742336 | 985 | 7535 | re-share of post | 2222.647625 | 6297.352375 | -26.251958 | 15.596172 |
| 16948544 | 36162 | 742025 | 985 | 7524 | this is re-share | 2219.778009 | 6289.221991 | -26.208014 | 15.570065 |
| 16718892 | 895 | 115784 | 1 | 1952 | taquin caroline taquin | 509.487184 | 1443.512816 | -22.527520 | 13.383500 |
| 3178384 | 895 | 115286 | 1 | 1947 | caroline taquin caroline | 508.182814 | 1439.817186 | -22.498551 | 13.366289 |
| 17607282 | 4116 | 232262 | 19 | 1998 | van der valk | 526.183129 | 1490.816871 | -22.110387 | 13.135683 |
| 12884750 | 0 | 134947 | 0 | 1812 | photo from commune | 472.703931 | 1339.296069 | -21.741755 | 12.916681 |
| 12885493 | 0 | 192240 | 0 | 1417 | photo from ville | 369.658648 | 1047.341352 | -19.226509 | 11.422384 |
| 514077 | 4 | 306789 | 1 | 1401 | 3 3 3 | 365.745536 | 1036.254464 | -19.072186 | 11.330701 |
| 11305102 | 0 | 142162 | 0 | 1307 | mp page fb | 340.962494 | 966.037506 | -18.465170 | 10.970075 |
| 6451200 | 0 | 141902 | 0 | 1305 | envoyer mp page | 340.440745 | 964.559255 | -18.451036 | 10.961679 |
| 18161030 | 17907 | 201183 | 25 | 1441 | vouloir prendre contact | 382.441481 | 1083.558519 | -18.277738 | 10.858723 |
| 2125639 | 1014 | 113137 | 47 | 1505 | auvio voir vidéo | 404.876657 | 1147.123343 | -17.785743 | 10.566431 |
| 16395658 | 1014 | 113091 | 47 | 1503 | streaming auvio voir | 404.354908 | 1145.645092 | -17.771267 | 10.557831 |
| 7938601 | 1014 | 112134 | 47 | 1476 | gratuitement vidéo programme | 397.311307 | 1125.688693 | -17.574731 | 10.441070 |
| 17979652 | 1041 | 112134 | 48 | 1476 | visionner gratuitement vidéo | 397.572181 | 1126.427819 | -17.531895 | 10.415621 |
| 10355426 | 0 | 153344 | 0 | 1138 | mamma africa news | 296.874765 | 841.125235 | -17.230054 | 10.236299 |
| 743543 | 0 | 152900 | 0 | 1137 | abonner vous mamma | 296.613891 | 840.386109 | -17.222482 | 10.231800 |
| 18185100 | 0 | 152892 | 0 | 1136 | vous mamma africa | 296.353017 | 839.646983 | -17.214907 | 10.227300 |
| 2125624 | 977 | 100542 | 46 | 1407 | auvio visionner gratuitement | 379.050117 | 1073.949883 | -17.106504 | 10.162898 |
| 1029707 | 0 | 150465 | 0 | 1117 | africa news info | 291.396408 | 825.603592 | -17.070337 | 10.141412 |
| 13470481 | 2542 | 883621 | 64 | 1476 | prendre soin proche | 401.746167 | 1138.253833 | -16.850569 | 10.010848 |
| 12884808 | 0 | 1155683 | 0 | 982 | photo from disneyland | 256.178400 | 725.821600 | -16.005574 | 9.508841 |
| 7787734 | 0 | 489648 | 0 | 968 | ginny carine anguima | 252.526162 | 715.473838 | -15.891072 | 9.440815 |
| 7538589 | 0 | 102368 | 0 | 961 | from voiture abandonner | 250.700043 | 710.299957 | -15.833510 | 9.406618 |
| 18090508 | 0 | 102368 | 0 | 961 | voiture abandonner monde' | 250.700043 | 710.299957 | -15.833510 | 9.406618 |
| 727761 | 0 | 102368 | 0 | 961 | abandonner monde' post | 250.700043 | 710.299957 | -15.833510 | 9.406618 |
| 12884659 | 0 | 81341 | 0 | 939 | photo from boutique | 244.960812 | 694.039188 | -15.651224 | 9.298323 |
| 2191922 | 0 | 127141 | 0 | 938 | avoir recueillir vouloir | 244.699938 | 693.300062 | -15.642888 | 9.293370 |
| 14451028 | 0 | 127206 | 0 | 938 | recueillir vouloir prendre | 244.699938 | 693.300062 | -15.642888 | 9.293370 |
| 13446978 | 4 | 99909 | 1 | 931 | prendre contact vite | 243.134693 | 688.865307 | -15.528645 | 9.225499 |
| 12885505 | 0 | 98787 | 0 | 923 | photo from voiture | 240.786826 | 682.213174 | -15.517307 | 9.218763 |
| 8489512 | 657 | 304517 | 5 | 949 | ici photo from | 248.873924 | 705.126076 | -15.458796 | 9.184002 |
| 6625469 | 0 | 478957 | 0 | 910 | ev ginny carine | 237.395462 | 672.604538 | -15.407643 | 9.153612 |
| 3178414 | 0 | 48366 | 0 | 880 | caroline taquin député | 229.569238 | 650.430762 | -15.151542 | 9.001464 |
| 16718954 | 0 | 46981 | 0 | 862 | taquin député bourgmestre | 224.873504 | 637.126496 | -14.995783 | 8.908928 |
| 5380191 | 0 | 1073806 | 0 | 801 | disneyland paris bon | 208.960182 | 592.039818 | -14.455455 | 8.587922 |
| 12441492 | 5406 | 113426 | 40 | 1044 | parti socialiste ps | 282.787562 | 801.212438 | -14.437642 | 8.577339 |
| 8656498 | 683 | 3323750 | 9 | 852 | in het nederlands | 224.612630 | 636.387370 | -14.386565 | 8.546994 |
| 12050652 | 115 | 3338395 | 5 | 826 | ook in het | 216.786406 | 614.213594 | -14.384079 | 8.545517 |
| 18115245 | 115 | 3342756 | 5 | 823 | volg ons ook | 216.003783 | 611.996217 | -14.356863 | 8.529349 |
| 12046128 | 115 | 3336266 | 5 | 821 | ons ook in | 215.482035 | 610.517965 | -14.338691 | 8.518553 |
| 8221272 | 115 | 3289625 | 5 | 810 | het nederlands op | 212.612419 | 602.387581 | -14.238329 | 8.458928 |
| 7537885 | 0 | 124176 | 0 | 774 | from poils moustaches' | 201.916580 | 572.083420 | -14.209735 | 8.441941 |
| 13118498 | 0 | 124176 | 0 | 774 | poils moustaches' post | 201.916580 | 572.083420 | -14.209735 | 8.441941 |
| 12266642 | 115 | 3231585 | 5 | 798 | pairi daiza nl | 209.481930 | 593.518070 | -14.128031 | 8.393401 |
| 12054369 | 115 | 3228050 | 5 | 794 | op pairi daiza | 208.438433 | 590.561567 | -14.091074 | 8.371445 |
fig = px.bar(graph_nonmedia3, x="res_nonmedia", y="trigramme", labels={"res_nonmedia":"Résiduel de Pearson","trigramme":"Trigramme (dans les posts Facebook <b>non-médias</b>)"}, orientation="h", title="Trigrammes les plus caractéristiques des posts Facebook <b>non-médias</b> {} en 2020".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="Olive", opacity=0.75, textfont_size=12)
fig
khi2_3["exp_inter_media"] = ((khi2_3.interactions_media + khi2_3.interactions_nonmedia) * khi2_3.interactions_media.sum()) / (khi2_3.interactions_media.sum() + khi2_3.interactions_nonmedia.sum())
khi2_3["exp_inter_nonmedia"] = ((khi2_3.interactions_media + khi2_3.interactions_nonmedia) * khi2_3.interactions_nonmedia.sum()) / (khi2_3.interactions_media.sum() + khi2_3.interactions_nonmedia.sum())
khi2_3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes colombe | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 15.834680 | 63.165320 |
| 1 | 0 | 6 | 0 | 1 | #2 continue faire | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 1.202634 | 4.797366 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle danse | 0.521748 | 1.478252 | -0.722321 | 0.429128 | 3.607902 | 14.392098 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife vivaforreyer | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 165.562607 | 660.437393 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique belgië | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 138.904221 | 554.095779 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 18981331 | 0 | 73 | 0 | 1 | single musique eden | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 14.632046 | 58.367954 |
| 18981332 | 0 | 12 | 0 | 1 | single streaming timeline | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.405268 | 9.594732 |
| 18981333 | 0 | 12 | 0 | 1 | solidaire en 2021 | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.405268 | 9.594732 |
| 18981334 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя parfum | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.405268 | 9.594732 |
| 18981335 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière temps | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.204829 | 8.795171 |
18981336 rows × 11 columns
khi2_3["res_inter_media"] = (khi2_3.interactions_media - khi2_3.exp_inter_media) / np.sqrt(khi2_3.exp_inter_media)
khi2_3["res_inter_nonmedia"] = (khi2_3.interactions_nonmedia - khi2_3.exp_inter_nonmedia) / np.sqrt(khi2_3.exp_inter_nonmedia)
khi2_3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 79 | 0 | 1 | #1 cafes colombe | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 15.834680 | 63.165320 | -3.979281 | 1.992370 |
| 1 | 0 | 6 | 0 | 1 | #2 continue faire | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 1.202634 | 4.797366 | -1.096647 | 0.549075 |
| 2 | 0 | 18 | 0 | 2 | #2 spectacle danse | 0.521748 | 1.478252 | -0.722321 | 0.429128 | 3.607902 | 14.392098 | -1.899448 | 0.951027 |
| 3 | 0 | 826 | 0 | 1 | #2020toussolidaire vivaforlife vivaforreyer | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 165.562607 | 660.437393 | -12.867113 | 6.442381 |
| 4 | 0 | 693 | 0 | 1 | #21juillet belgique belgië | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 138.904221 | 554.095779 | -11.785764 | 5.900965 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 18981331 | 0 | 73 | 0 | 1 | single musique eden | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 14.632046 | 58.367954 | -3.825186 | 1.915216 |
| 18981332 | 0 | 12 | 0 | 1 | single streaming timeline | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.405268 | 9.594732 | -1.550893 | 0.776510 |
| 18981333 | 0 | 12 | 0 | 1 | solidaire en 2021 | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.405268 | 9.594732 | -1.550893 | 0.776510 |
| 18981334 | 0 | 12 | 0 | 1 | 𐀁мвя иσιя parfum | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.405268 | 9.594732 | -1.550893 | 0.776510 |
| 18981335 | 0 | 11 | 0 | 1 | 𐌾ᗩïᗩᙅ lumière temps | 0.260874 | 0.739126 | -0.510758 | 0.303439 | 2.204829 | 8.795171 | -1.484867 | 0.743452 |
18981336 rows × 13 columns
graph_interactions_media3 = khi2_3.sort_values(by=["res_inter_media"],ascending=False).head(50)
graph_interactions_media3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 11379766 | 713970 | 1157 | 4 | 2 | mère fils 15 | 1.565245 | 4.434755 | 1.946096 | -1.156167 | 143339.334986 | 5.717877e+05 | 1507.205304 | -754.636340 |
| 7149196 | 714035 | 1710 | 6 | 4 | fils 15 an | 2.608741 | 7.391259 | 2.099643 | -1.247388 | 143463.206283 | 5.722818e+05 | 1506.399046 | -754.232659 |
| 4321599 | 1003102 | 559225 | 3831 | 3824 | conseil national sécurité | 1996.991498 | 5658.008502 | 41.040556 | -24.382012 | 313151.248954 | 1.249176e+06 | 1232.937374 | -617.314274 |
| 8436613 | 433985 | 1950 | 5 | 3 | héberger ami besoin | 2.086993 | 5.913007 | 2.016420 | -1.197946 | 87378.371950 | 3.485566e+05 | 1172.559498 | -587.083927 |
| 16019389 | 408314 | 3903 | 27 | 6 | soin intensif cause | 8.608846 | 24.391154 | 6.268108 | -3.723855 | 82624.359939 | 3.295926e+05 | 1133.052126 | -567.303146 |
| 8905166 | 387128 | 1599 | 15 | 4 | intensif cause coronavirus | 4.956609 | 14.043391 | 4.511158 | -2.680059 | 77916.048018 | 3.108110e+05 | 1107.752692 | -554.636077 |
| 1413961 | 379157 | 3328 | 19 | 10 | an soin intensif | 7.565350 | 21.434650 | 4.157271 | -2.469816 | 76664.907831 | 3.058201e+05 | 1092.485589 | -546.992055 |
| 2900193 | 359272 | 73 | 4 | 2 | bruxelles contracter covid-19 | 1.565245 | 4.434755 | 1.946096 | -1.156167 | 72026.749557 | 2.873183e+05 | 1070.301042 | -535.884567 |
| 16025965 | 356985 | 0 | 2 | 0 | soin témoignage mère | 0.521748 | 1.478252 | 2.046529 | -1.215834 | 71553.713536 | 2.854313e+05 | 1067.051759 | -534.257697 |
| 13470909 | 356985 | 0 | 2 | 0 | prendre soin témoignage | 0.521748 | 1.478252 | 2.046529 | -1.215834 | 71553.713536 | 2.854313e+05 | 1067.051759 | -534.257697 |
| 8460621 | 357060 | 92 | 4 | 3 | hôpital reine fabiola | 1.826119 | 5.173881 | 1.608686 | -0.955713 | 71587.186848 | 2.855648e+05 | 1066.957466 | -534.210485 |
| 3580558 | 356985 | 55 | 2 | 1 | choquet mère fils | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 10827270 | 356985 | 55 | 2 | 1 | mettre masque plaire | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 13020477 | 356985 | 55 | 2 | 1 | plaire appel déchirant | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 1419574 | 356985 | 55 | 2 | 1 | an venir transférer | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 5716597 | 356985 | 55 | 2 | 1 | déchirant françoise choquet | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 1653846 | 356985 | 55 | 2 | 1 | appel déchirant françoise | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 10548731 | 356985 | 55 | 2 | 1 | masque plaire appel | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 7509023 | 356985 | 55 | 2 | 1 | françoise choquet mère | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 17185136 | 356985 | 55 | 2 | 1 | transférer hôpital reine | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 6776722 | 356985 | 55 | 2 | 1 | fabiola bruxelles contracter | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 4445962 | 356985 | 55 | 2 | 1 | contracter covid-19 prendre | 0.782622 | 2.217378 | 1.376097 | -0.817533 | 71564.737681 | 2.854753e+05 | 1066.928360 | -534.195912 |
| 17706313 | 356985 | 75 | 2 | 2 | venir transférer hôpital | 1.043497 | 2.956503 | 0.936356 | -0.556285 | 71568.746461 | 2.854913e+05 | 1066.883493 | -534.173449 |
| 171019 | 357024 | 511 | 3 | 5 | 15 an venir | 2.086993 | 5.913007 | 0.631995 | -0.375465 | 71663.954982 | 2.858710e+05 | 1065.964591 | -533.713367 |
| 14500059 | 356985 | 587 | 2 | 8 | reine fabiola bruxelles | 2.608741 | 7.391259 | -0.376892 | 0.223910 | 71671.371225 | 2.859006e+05 | 1065.736060 | -533.598945 |
| 170940 | 356985 | 1107 | 2 | 2 | 15 an soin | 1.043497 | 2.956503 | 0.936356 | -0.556285 | 71775.599500 | 2.863164e+05 | 1064.572937 | -533.016586 |
| 17405752 | 356985 | 1388 | 2 | 2 | témoignage mère fils | 1.043497 | 2.956503 | 0.936356 | -0.556285 | 71831.922857 | 2.865411e+05 | 1063.945339 | -532.702356 |
| 4724673 | 356985 | 6402 | 2 | 34 | covid-19 prendre soin | 9.391469 | 26.608531 | -2.411926 | 1.432915 | 72836.923963 | 2.905501e+05 | 1052.855867 | -527.150015 |
| 18399250 | 334820 | 15156 | 789 | 55 | wout van aert | 220.177769 | 623.822231 | 38.334502 | -22.774358 | 70148.836642 | 2.798272e+05 | 999.301156 | -500.335930 |
| 15022690 | 312212 | 0 | 433 | 0 | rtl info' post | 112.958500 | 320.041500 | 30.112502 | -17.889704 | 62579.458550 | 2.496325e+05 | 997.896036 | -499.632406 |
| 7538015 | 312212 | 0 | 433 | 0 | from rtl info' | 112.958500 | 320.041500 | 30.112502 | -17.889704 | 62579.458550 | 2.496325e+05 | 997.896036 | -499.632406 |
| 10421499 | 383013 | 58659 | 1149 | 236 | marc van ranst | 361.310676 | 1023.689324 | 41.439505 | -24.619026 | 88528.290446 | 3.531437e+05 | 989.741341 | -495.549466 |
| 17831670 | 325573 | 36982 | 9 | 12 | vie annie cordy | 5.478357 | 15.521643 | 1.504597 | -0.893874 | 72670.158721 | 2.898848e+05 | 938.157282 | -469.722059 |
| 1507969 | 346571 | 61213 | 60 | 53 | annie cordy décéder | 29.478777 | 83.521223 | 5.621435 | -3.339670 | 81735.813888 | 3.260482e+05 | 926.338020 | -463.804322 |
| 12885351 | 279752 | 7859 | 380 | 13 | photo from rtl | 102.523535 | 290.476465 | 27.404026 | -16.280610 | 57648.458910 | 2.299625e+05 | 925.042384 | -463.155615 |
| 15021172 | 321975 | 42982 | 1533 | 523 | rtc télé liège | 536.357220 | 1519.642780 | 43.034074 | -25.566352 | 73151.613180 | 2.918054e+05 | 919.981816 | -460.621860 |
| 8887323 | 280842 | 14583 | 924 | 40 | institut santé public | 251.482666 | 712.517334 | 42.408163 | -25.194501 | 59214.689193 | 2.362103e+05 | 910.769650 | -456.009459 |
| 5122140 | 254875 | 8005 | 1116 | 160 | dernier 24 heure | 332.875395 | 943.124605 | 42.922995 | -25.500361 | 52691.402200 | 2.101886e+05 | 880.797437 | -441.002797 |
| 18720987 | 237912 | 4790 | 683 | 50 | épidémie coronavirus belgique | 191.220740 | 541.779260 | 35.563354 | -21.128031 | 48646.944221 | 1.940551e+05 | 858.109390 | -429.643213 |
| 18494706 | 352747 | 111437 | 143 | 134 | âge 92 an | 72.262135 | 204.737865 | 8.321403 | -4.943709 | 93040.573033 | 3.711434e+05 | 851.425218 | -426.296543 |
| 1394768 | 237861 | 10719 | 10 | 8 | an julie melissa | 4.695734 | 13.304266 | 2.447786 | -1.454219 | 49825.124615 | 1.987549e+05 | 842.396433 | -421.775958 |
| 17941739 | 219809 | 991 | 6 | 2 | virginie dénoncer voisin | 2.086993 | 5.913007 | 2.708633 | -1.609187 | 44256.929419 | 1.765431e+05 | 834.478817 | -417.811720 |
| 5933956 | 219454 | 944 | 3 | 1 | dénoncer voisin héberger | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 8968315 | 219454 | 944 | 3 | 1 | intervention policier suivre | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 16483244 | 219454 | 944 | 3 | 1 | suivre profondément marquer | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 1304788 | 219454 | 944 | 3 | 1 | ami besoin carolo | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 5933958 | 219454 | 944 | 3 | 1 | dénoncer voisin intervention | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 18087774 | 219454 | 944 | 3 | 1 | voisin intervention policier | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 2509420 | 219454 | 944 | 3 | 1 | besoin carolo dénoncer | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
| 3179108 | 219454 | 944 | 3 | 1 | carolo dénoncer voisin | 1.043497 | 2.956503 | 1.915293 | -1.137867 | 44176.352945 | 1.762216e+05 | 833.933855 | -417.538865 |
fig = px.bar(graph_interactions_media3, x="res_inter_media", y="trigramme", labels={"res_inter_media":"Résiduel de Pearson (<b>en fonction des interactions</b>)","trigramme":"Trigramme (dans les posts Facebook <b>médias</b>)"}, orientation="h", title="Trigrammes les plus caractéristiques des posts Facebook <b>médias</b> {} en 2020<br>(en fonction des interactions suscitées)".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="green", opacity=0.75, textfont_size=12)
fig
graph_interactions_nonmedia3 = khi2_3.sort_values(by=["res_inter_nonmedia"],ascending=False).head(50)
graph_interactions_nonmedia3
| interactions_media | interactions_nonmedia | obs_media | obs_nonmedia | trigramme | exp_media | exp_nonmedia | res_media | res_nonmedia | exp_inter_media | exp_inter_nonmedia | res_inter_media | res_inter_nonmedia | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 16935230 | 54332 | 4207760 | 181 | 596 | the voice kids | 202.699202 | 574.300798 | -1.524115 | 0.905470 | 854289.424018 | 3.407803e+06 | -865.494591 | 433.340879 |
| 18115245 | 115 | 3342756 | 5 | 823 | volg ons ook | 216.003783 | 611.996217 | -14.356863 | 8.529349 | 670041.693412 | 2.672829e+06 | -818.420255 | 409.771425 |
| 12050652 | 115 | 3338395 | 5 | 826 | ook in het | 216.786406 | 614.213594 | -14.384079 | 8.545517 | 669167.578968 | 2.669342e+06 | -817.886055 | 409.503959 |
| 12046128 | 115 | 3336266 | 5 | 821 | ons ook in | 215.482035 | 610.517965 | -14.338691 | 8.518553 | 668740.844354 | 2.667640e+06 | -817.625137 | 409.373321 |
| 8656498 | 683 | 3323750 | 9 | 852 | in het nederlands | 224.612630 | 636.387370 | -14.386565 | 8.546994 | 666345.999279 | 2.658087e+06 | -815.463487 | 408.291014 |
| 8221272 | 115 | 3289625 | 5 | 810 | het nederlands op | 212.612419 | 602.387581 | -14.238329 | 8.458928 | 659392.169332 | 2.630348e+06 | -811.888040 | 406.500838 |
| 12266642 | 115 | 3231585 | 5 | 798 | pairi daiza nl | 209.481930 | 593.518070 | -14.128031 | 8.393401 | 647758.690240 | 2.583941e+06 | -804.691687 | 402.897726 |
| 12054369 | 115 | 3228050 | 5 | 794 | op pairi daiza | 208.438433 | 590.561567 | -14.091074 | 8.371445 | 647050.138403 | 2.581115e+06 | -804.251303 | 402.677232 |
| 11538468 | 115 | 3203780 | 5 | 789 | nederlands op pairi | 207.134063 | 586.865937 | -14.044741 | 8.343918 | 642185.484069 | 2.561710e+06 | -801.221258 | 401.160132 |
| 12433340 | 0 | 3118599 | 0 | 34 | partenaire touré collection | 8.869721 | 25.130279 | -2.978208 | 1.769340 | 625088.839813 | 2.493510e+06 | -790.625600 | 395.855036 |
| 12266932 | 0 | 2870770 | 0 | 754 | pairi daiza' post | 196.699097 | 557.300903 | -14.024946 | 8.332158 | 575414.244880 | 2.295356e+06 | -758.560640 | 379.800565 |
| 7537804 | 0 | 2870770 | 0 | 754 | from pairi daiza' | 196.699097 | 557.300903 | -14.024946 | 8.332158 | 575414.244880 | 2.295356e+06 | -758.560640 | 379.800565 |
| 12885242 | 0 | 2845866 | 0 | 749 | photo from pairi | 195.394727 | 553.605273 | -13.978366 | 8.304485 | 570422.512225 | 2.275443e+06 | -755.263207 | 378.149587 |
| 3874650 | 0 | 2700135 | 0 | 30 | collection likeetpartage cesttouss | 7.826224 | 22.173776 | -2.797539 | 1.662005 | 541212.337491 | 2.158923e+06 | -735.671352 | 368.340222 |
| 17128108 | 0 | 2699866 | 0 | 29 | touré collection likeetpartage | 7.565350 | 21.434650 | -2.750518 | 1.634071 | 541158.419402 | 2.158708e+06 | -735.634705 | 368.321874 |
| 11616691 | 0 | 2263822 | 0 | 617 | nl photo from | 160.959341 | 456.040659 | -12.686975 | 7.537276 | 453758.199603 | 1.810064e+06 | -673.615766 | 337.269869 |
| 4978524 | 0 | 2262072 | 0 | 598 | daiza nl photo | 156.002732 | 441.997268 | -12.490105 | 7.420316 | 453407.431367 | 1.808665e+06 | -673.355353 | 337.139485 |
| 4524158 | 0 | 1768456 | 0 | 44 | coronavirus confinement day | 11.478462 | 32.521538 | -3.387988 | 2.012789 | 354467.537923 | 1.413988e+06 | -595.371764 | 298.094207 |
| 12674579 | 0 | 1550146 | 0 | 270 | pensée grégory 3 | 70.436016 | 199.563984 | -8.392617 | 4.986016 | 310709.701594 | 1.239436e+06 | -557.413403 | 279.088993 |
| 1301993 | 0 | 1542369 | 0 | 10 | america' got talent | 2.608741 | 7.391259 | -1.615160 | 0.959559 | 309150.887553 | 1.233218e+06 | -556.013388 | 278.388025 |
| 9474140 | 0 | 1542327 | 0 | 5 | jury america' got | 1.304371 | 3.695629 | -1.142090 | 0.678511 | 309142.469116 | 1.233185e+06 | -556.005818 | 278.384235 |
| 8867994 | 0 | 1470408 | 0 | 15 | instagram partenaire touré | 3.913112 | 11.086888 | -1.978159 | 1.175215 | 294727.097255 | 1.175681e+06 | -542.887739 | 271.816199 |
| 2731876 | 0 | 1435319 | 0 | 264 | bonjour souhaite bon | 68.870771 | 195.129229 | -8.298842 | 4.930305 | 287693.893467 | 1.147625e+06 | -536.371041 | 268.553380 |
| 18423005 | 0 | 1372935 | 0 | 13 | yao pawa gang | 3.391364 | 9.608636 | -1.841566 | 1.094066 | 275189.707394 | 1.097745e+06 | -524.585272 | 262.652413 |
| 12254282 | 5537 | 1353910 | 113 | 324 | page météo mons | 114.001997 | 322.998003 | -0.093845 | 0.055753 | 272486.186270 | 1.086961e+06 | -511.394857 | 256.048159 |
| 12253988 | 0 | 1263899 | 0 | 147 | page merveill monde | 38.348498 | 108.651502 | -6.192616 | 3.679006 | 253334.641469 | 1.010564e+06 | -503.323595 | 252.006993 |
| 8869151 | 0 | 1260098 | 0 | 219 | instagram tweeter samuelnito5 | 57.131435 | 161.868565 | -7.558534 | 4.490491 | 252572.772861 | 1.007525e+06 | -502.566188 | 251.627769 |
| 514804 | 0 | 1196944 | 0 | 240 | 3 bisous 3 | 62.609792 | 177.390208 | -7.912635 | 4.700861 | 239914.248764 | 9.570298e+05 | -489.810421 | 245.241138 |
| 11122716 | 0 | 1167195 | 0 | 157 | monde dicton jour | 40.957239 | 116.042761 | -6.399784 | 3.802083 | 233951.389193 | 9.332436e+05 | -483.685217 | 242.174335 |
| 6573194 | 0 | 1157241 | 0 | 11 | eudoxie yao pawa | 2.869615 | 8.130385 | -1.693994 | 1.006394 | 231956.219467 | 9.252848e+05 | -481.618334 | 241.139476 |
| 8014706 | 0 | 1155958 | 0 | 233 | grégory 3 bisous | 60.783673 | 172.216327 | -7.796388 | 4.631800 | 231699.056240 | 9.242589e+05 | -481.351282 | 241.005766 |
| 12884808 | 0 | 1155683 | 0 | 982 | photo from disneyland | 256.178400 | 725.821600 | -16.005574 | 9.508841 | 231643.935518 | 9.240391e+05 | -481.294022 | 240.977097 |
| 516727 | 0 | 1135904 | 0 | 140 | 3 pensée grégory | 36.522379 | 103.477621 | -6.043375 | 3.590342 | 227679.452696 | 9.082245e+05 | -477.157681 | 238.906090 |
| 16484010 | 0 | 1096392 | 0 | 227 | suivre réseau social | 59.218428 | 167.781572 | -7.695351 | 4.571774 | 219759.707247 | 8.766323e+05 | -468.785353 | 234.714184 |
| 134874 | 0 | 1079127 | 0 | 3 | 13 an impressionne | 0.782622 | 2.217378 | -0.884659 | 0.525572 | 216299.128051 | 8.628279e+05 | -465.079701 | 232.858817 |
| 1392041 | 0 | 1079127 | 0 | 3 | an impressionne jury | 0.782622 | 2.217378 | -0.884659 | 0.525572 | 216299.128051 | 8.628279e+05 | -465.079701 | 232.858817 |
| 8641847 | 0 | 1079127 | 0 | 3 | impressionne jury america' | 0.782622 | 2.217378 | -0.884659 | 0.525572 | 216299.128051 | 8.628279e+05 | -465.079701 | 232.858817 |
| 7829680 | 0 | 1079127 | 0 | 3 | got talent bohemian | 0.782622 | 2.217378 | -0.884659 | 0.525572 | 216299.128051 | 8.628279e+05 | -465.079701 | 232.858817 |
| 17369370 | 0 | 1077628 | 0 | 172 | tweeter samuelnito5 lapersévérenceestaurendezvous | 44.870351 | 127.129649 | -6.698534 | 3.979569 | 215998.670002 | 8.616293e+05 | -464.756571 | 232.697030 |
| 5380191 | 0 | 1073806 | 0 | 801 | disneyland paris bon | 208.960182 | 592.039818 | -14.455455 | 8.587922 | 215232.592175 | 8.585734e+05 | -463.931668 | 232.284013 |
| 18390204 | 0 | 1062361 | 0 | 337 | woman and material | 87.914583 | 249.085417 | -9.376278 | 5.570405 | 212938.567912 | 8.494224e+05 | -461.452671 | 231.042814 |
| 3568988 | 0 | 1043704 | 0 | 15 | choisir réponse ici | 3.913112 | 11.086888 | -1.978159 | 1.175215 | 209198.977640 | 8.345050e+05 | -457.382747 | 229.005061 |
| 18442374 | 0 | 1024798 | 0 | 175 | youtube channel ucefx7o81rsf6-z6sufthoxa | 45.652974 | 129.347026 | -6.756698 | 4.014125 | 205409.478058 | 8.193885e+05 | -453.221224 | 226.921445 |
| 3395864 | 0 | 1012130 | 0 | 171 | channel ucefx7o81rsf6-z6sufthoxa instagram | 44.609477 | 126.390523 | -6.679033 | 3.967984 | 202870.316908 | 8.092597e+05 | -450.411275 | 225.514544 |
| 17430581 | 0 | 1009648 | 0 | 162 | ucefx7o81rsf6-z6sufthoxa instagram tweeter | 42.261610 | 119.738390 | -6.500893 | 3.862152 | 202372.827330 | 8.072752e+05 | -449.858675 | 225.237865 |
| 7536897 | 0 | 1009227 | 0 | 626 | from disneyland paris | 163.307208 | 462.692792 | -12.779171 | 7.592049 | 202288.442515 | 8.069386e+05 | -449.764875 | 225.190900 |
| 12365916 | 0 | 1008497 | 0 | 625 | paris bon plans' | 163.046334 | 461.953666 | -12.768960 | 7.585982 | 202142.122050 | 8.063549e+05 | -449.602182 | 225.109442 |
| 2706826 | 0 | 1008497 | 0 | 625 | bon plans' post | 163.046334 | 461.953666 | -12.768960 | 7.585982 | 202142.122050 | 8.063549e+05 | -449.602182 | 225.109442 |
| 5564676 | 0 | 1006727 | 0 | 11 | dougourouni tv instagram | 2.869615 | 8.130385 | -1.693994 | 1.006394 | 201787.345035 | 8.049397e+05 | -449.207463 | 224.911812 |
| 10884068 | 0 | 999560 | 0 | 6 | mieux ecouter coeur' | 1.565245 | 4.434755 | -1.251097 | 0.743271 | 200350.798780 | 7.992092e+05 | -447.605629 | 224.109796 |
fig = px.bar(graph_interactions_nonmedia3, x="res_inter_nonmedia", y="trigramme", labels={"res_inter_nonmedia":"Résiduel de Pearson (<b>en fonction des interactions</b>)","trigramme":"Trigramme (dans les posts Facebook <b>non-médias</b>)"}, orientation="h", title="Trigrammes les plus caractéristiques des posts Facebook <b>non-médias</b> {} en 2020<br>(en fonction des interactions suscitées)".format(pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=1000)
fig.update_traces(marker_color="lightgreen", opacity=0.75, textfont_size=12)
fig
pays = "the French Community of Belgium"
termeMaj = "Lemma"
termeMin = "lemma"
yy = "mot"
media = "media"
source = graph_interactions_media1[:10]
xx= "res_inter_media"
fig = px.bar(source, x=xx, y=yy, labels={xx:"χ<sup>2</sup> residual (<b>weighed by interactions</b>)",yy:"{} (in <b>{}</b> Facebook posts)".format(termeMaj,media)}, orientation="h", title="Most characteristic {}s among <b>{}</b> Facebook posts in {} (2020)".format(termeMin,media,pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=400)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig
termeMaj = "Lemma"
termeMin = "lemma"
yy = "mot"
media = "non-media"
source = graph_interactions_nonmedia1[:10]
xx= "res_inter_nonmedia"
fig = px.bar(source, x=xx, y=yy, labels={xx:"χ<sup>2</sup> residual (<b>weighed by interactions</b>)",yy:"{} (in <b>{}</b> Facebook posts)".format(termeMaj,media)}, orientation="h", title="Most characteristic {}s among <b>{}</b> Facebook posts in {} (2020)".format(termeMin,media,pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=400)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig
termeMaj = "Bigram"
termeMin = "bigram"
yy = "bigramme"
media = "media"
source = graph_interactions_media2[:10]
xx= "res_inter_media"
fig = px.bar(source, x=xx, y=yy, labels={xx:"χ<sup>2</sup> residual (<b>weighed by interactions</b>)",yy:"{} (in <b>{}</b> Facebook posts)".format(termeMaj,media)}, orientation="h", title="Most characteristic {}s among <b>{}</b> Facebook posts in {} (2020)".format(termeMin,media,pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=400)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig
termeMaj = "Bigram"
termeMin = "bigram"
yy = "bigramme"
media = "non-media"
source = graph_interactions_nonmedia2[:10]
xx= "res_inter_nonmedia"
fig = px.bar(source, x=xx, y=yy, labels={xx:"χ<sup>2</sup> residual (<b>weighed by interactions</b>)",yy:"{} (in <b>{}</b> Facebook posts)".format(termeMaj,media)}, orientation="h", title="Most characteristic {}s among <b>{}</b> Facebook posts in {} (2020)".format(termeMin,media,pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=400)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig
termeMaj = "Trigram"
termeMin = "trigram"
yy = "trigramme"
media = "media"
source = graph_interactions_media3[:10]
xx= "res_inter_media"
fig = px.bar(source, x=xx, y=yy, labels={xx:"χ<sup>2</sup> residual (<b>weighed by interactions</b>)",yy:"{} (in <b>{}</b> Facebook posts)".format(termeMaj,media)}, orientation="h", title="Most characteristic {}s among <b>{}</b> Facebook posts in {} (2020)".format(termeMin,media,pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=400)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig
termeMaj = "Trigram"
termeMin = "trigram"
yy = "trigramme"
media = "non-media"
source = graph_interactions_nonmedia3[:10]
xx = "res_inter_nonmedia"
fig = px.bar(source, x=xx, y=yy, labels={xx:"χ<sup>2</sup> residual (<b>weighed by interactions</b>)",yy:"{} (in <b>{}</b> Facebook posts)".format(termeMaj,media)}, orientation="h", title="Most characteristic {}s among <b>{}</b> Facebook posts in {} (2020)".format(termeMin,media,pays))
fig.update_layout(yaxis=dict(autorange="reversed"), width=1000, height=400)
fig.update_traces(marker_color="blue", opacity=0.75, textfont_size=12)
fig