-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy path.Rhistory
512 lines (512 loc) · 26.1 KB
/
.Rhistory
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
SOTU_fre_sent <- data.frame("year" = SOTU_year, "SOTU_stat" = SOTU_stat$meanSentenceLength)
SOTU_fre_word <- data.frame("year" = SOTU_year, "SOTU_stat" = SOTU_stat$meanWordSyllables)
###EO
data("data_corpus_eo")
#drop the ones that are empty
clean1 <- data_corpus_eo$documents
##problems
clean1$texts<-stri_replace_all_fixed(clean1$texts, "A. D.", "AD")
#drop anything that's very long (95th percentile and above)
quant95 <- which( nchar(clean1$texts) >
quantile( nchar(clean1$texts), prob=.95) )
clean2 <- clean1[-quant95, c(1:2)]
eo_corp <- corpus(clean2, text_field = "texts")
##drop very short sentences
eo_corp <- corpus_trimsentences(eo_corp, min_length = 4)
eo_corp <- corpus_subset(eo_corp, ntoken(eo_corp) > 10)
eo_year <- docvars(eo_corp)
eo_stat <- textstat_readability(eo_corp$documents$texts, measure = c("Flesch", "meanSentenceLength", "meanWordSyllables"))
eo_fre_df <- data.frame("year" = eo_year$Year, "eo_stat" = eo_stat$Flesch)
eo_fre_sent <- data.frame("year" = eo_year$Year, "eo_stat" = eo_stat$meanSentenceLength)
eo_fre_word <- data.frame("year" = eo_year$Year, "eo_stat" = eo_stat$meanWordSyllables)
########SCOTUS
#load("C:/Users/kevin/Documents/GitHub/BMS_chapter_replication/data/data_corpus_SCOTUS.rda")
load("C:/Users/kevin/Desktop/data_corpus_SCOTUS.rda")
#drop the ones that are empty
clean1 <- corpus_subset(data_corpus_SCOTUS, nchar(texts(data_corpus_SCOTUS)) > 0)
## implement fixes
texts(clean1) <- stri_replace_all_fixed(texts(clean1), "U. S.", "US")
#drop anything that's very long (95th percentile and above)
temp_lengths <- stri_length(texts(clean1))
clean2 <- corpus_subset(clean1, temp_lengths <quantile(temp_lengths, prob = .95))
scotus_lengths<-ntoken(clean2, removePunct = T)
clean3 <- corpus_trimsentences(clean2, min_length = 4)
scotus_year <- clean3$documents$Year
scotus_stat <- textstat_readability(clean3$documents$texts, measure = c("Flesch", "meanSentenceLength", "meanWordSyllables"))
scotus_fre_df <- data.frame("year" = scotus_year, "scotus_stat" = scotus_stat$Flesch)
scotus_fre_sent <- data.frame("year" = scotus_year, "scotus_stat" = scotus_stat$meanSentenceLength)
scotus_fre_word <- data.frame("year" = scotus_year, "scotus_stat" = scotus_stat$meanWordSyllables)
##too many to plot well, and very unbalanced (way more modern ones)--try for a more balanced sample
years<-unique(scotus_fre_df$year)
indices<-vector()
for(i in 1:length(years)){
set<-which(scotus_fre_df[,"year"]==years[i])
num<-min(length(set), 30)
samp<-sample(set, num, replace = FALSE)
indices<-c(indices, samp)
}
balanced_scotus_fre_df<-scotus_fre_df[indices,]
balanced_scotus_fre_sent<-scotus_fre_sent[indices,]
balanced_scotus_fre_word<-scotus_fre_word[indices,]
require(reshape2)
###Congress
load("C:/Users/kevin/Desktop/data_corpus_CR.rdata")
#drop the ones that are by the speaker
speaker <- which( (data_corpus_CR$documents$name)=="Speaker" )
clean1 <- data_corpus_CR$documents[-speaker, ]
#drop anything that's very long (90th percentile and above)
quant95 <- which( nchar(clean1$texts) >
quantile( nchar(clean1$texts), prob=.95) )
clean2 <- clean1[-quant95, ]
##sync up year format
clean2$year[clean2$year==95]<-1995
clean2$year[clean2$year==96]<-1996
clean2$year[clean2$year==97]<-1997
clean2$year[clean2$year==98]<-1998
clean2$year[clean2$year==99]<-1999
clean2$year[clean2$year==0]<-2000
clean2$year[clean2$year==1]<-2001
clean2$year[clean2$year==2]<-2002
clean2$year[clean2$year==3]<-2003
clean2$year[clean2$year==4]<-2004
clean2$year[clean2$year==5]<-2005
clean2$year[clean2$year==6]<-2006
clean2$year[clean2$year==7]<-2007
clean2$year[clean2$year==8]<-2008
##take a very small sample--can't run readability on the whole thing
sample<-sample(seq(1,length(clean2$texts)), 10000, replace = F )
sample_data<-clean2[sample,]
congress_year <- sample_data$year
congress_stat <- textstat_readability(sample_data$texts, measure = c("Flesch", "meanSentenceLength", "meanWordSyllables"))
congress_fre_df <- data.frame("year" = congress_year, "congress_stat" = congress_stat$Flesch)
congress_fre_sent <- data.frame("year" = congress_year, "congress_stat" = congress_stat$meanSentenceLength)
congress_fre_word <- data.frame("year" = congress_year, "congress_stat" = congress_stat$meanWordSyllables)
##melt together for plotting--Figure 3
require(reshape2)
df_US<-melt(list(SOTU=SOTU_fre_df, SCOTUS=balanced_scotus_fre_df, Congress = congress_fre_df,
ExecOrders = eo_fre_df), id.vars="year")
require(ggplot2)
linez<-c( "F1", "dashed", "dotdash","solid")
p <- ggplot(data = df_US,
aes(x = year, y = value, linetype = L1)) + #, group = delivery)) +
theme(panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black")) +
geom_smooth(alpha=0.2, method = "loess", span = .34, color = "black", se = F) +
coord_cartesian(ylim = c(-20, 65)) +
theme(legend.position="none", axis.text.x = element_text(size = 15),
axis.text.y = element_text(size = 15), axis.title.y = element_text(size = 15))+
scale_linetype_manual(values = linez)+
xlab("") +
ylab("Flesch Reading Ease Score") +
theme(plot.title = element_text(lineheight=.8, face="bold")) +
annotate("text", label = "Congress Speech", x = 1995, y = 45, size = 6, colour = "black")+
annotate("text", label = "SOTU", x = 1800, y = 20, size = 6, colour = "black")+
annotate("text", label = "SCOTUS", x = 1810, y = 50, size = 6, colour = "black") +
annotate("text", label = "Executive Orders", x = 1970, y = 0, size = 6, colour = "black")
print(p)
##melt together for plotting--Figure 5
df_US_sent<-melt(list(SOTU=SOTU_fre_sent, SCOTUS=balanced_scotus_fre_sent, Congress = congress_fre_sent,
ExecOrders = eo_fre_sent), id.vars="year")
linez<-c( "F1", "dashed", "dotdash","solid")
p <- ggplot(data = df_US_sent,
aes(x = year, y = value, linetype = L1)) + #, group = delivery)) +
theme(panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black")) +
geom_smooth(alpha=0.2, method = "loess", span = .34, color = "black", se = F) +
coord_cartesian(ylim = c(15, 65)) +
theme(legend.position="none", axis.text.x = element_text(size = 15),
axis.text.y = element_text(size = 15), axis.title.y = element_text(size = 15))+
scale_linetype_manual(values = linez)+
xlab("") +
ylab("Mean Sentence Length") +
theme(plot.title = element_text(lineheight=.8, face="bold")) +
annotate("text", label = "Congress Speech", x = 1960, y = 17, size = 6, colour = "black")+
annotate("text", label = "SOTU", x = 1910, y = 25, size = 6, colour = "black")+
annotate("text", label = "SCOTUS", x = 1810, y = 30, size = 6, colour = "black") +
annotate("text", label = "Executive Orders", x = 1960, y = 50, size = 6, colour = "black")
print(p)
##melt together for plotting--Figure 7
df_US_word<-melt(list(SOTU=SOTU_fre_word, SCOTUS=balanced_scotus_fre_word, Congress = congress_fre_word,
ExecOrders = eo_fre_word), id.vars="year")
p <- ggplot(data = df_US_word,
aes(x = year, y = value, linetype = L1)) + #, group = delivery)) +
theme(panel.grid.major = element_blank(),
panel.grid.minor = element_blank(),
panel.background = element_blank(),
axis.line = element_line(colour = "black")) +
geom_smooth(alpha=0.2, method = "loess", span = .34, color = "black", se = F) +
coord_cartesian(ylim = c(1.4, 1.85)) +
theme(legend.position="none", axis.text.x = element_text(size = 15),
axis.text.y = element_text(size = 15), axis.title.y = element_text(size = 15))+
scale_linetype_manual(values = linez)+
xlab("") +
ylab("Mean Number of Syllables per Word") + theme(plot.title = element_text(lineheight=.8, face="bold")) +
annotate("text", label = "SOTU", x = 1800, y = 1.73, size = 6, colour = "black")+
annotate("text", label = "SCOTUS", x = 1810, y = 1.47, size = 6, colour = "black") +
annotate("text", label = "Con. Speech", x = 1998, y = 1.61, size = 6, colour = "black")+
annotate("text", label = "Executive Orders", x = 1950, y = 1.82, size = 6, colour = "black")
print(p)
##Figures 4, 6 and 8: Comparative data
data("data_corpus_SOTU")
SOTU_stat <- textstat_readability(data_corpus_SOTU, measure = c("Flesch", "meanSentenceLength", "meanWordSyllables"))
SOTU_year <- lubridate::year(docvars(data_corpus_SOTU, "Date"))
SOTU_fre_df <- data.frame("year" = SOTU_year, "SOTU_stat" = SOTU_stat$Flesch)
SOTU_fre_sent <- data.frame("year" = SOTU_year, "SOTU_stat" = SOTU_stat$meanSentenceLength)
SOTU_fre_word <- data.frame("year" = SOTU_year, "SOTU_stat" = SOTU_stat$meanWordSyllables)
########nobel
#load("data_text/NobelLitePresentations/data_corpus_nobel.rdata")
load("C:/Users/kevin/Desktop/data_corpus_nobel.rdata")
temp_lengths <- stri_length(texts(data_corpus_nobel))
data_corpus_nobel <- corpus_subset(data_corpus_nobel, temp_lengths < quantile(temp_lengths, prob = .95))
nobel_corp <- corpus_trimsentences(data_corpus_nobel, min_length = 4)
nobel_lengths<-ntoken(nobel_corp, removePunct = T)
nobel_stat <- textstat_readability(nobel_corp, measure = c("Flesch", "meanSentenceLength", "meanWordSyllables"))
nobel_year <- lubridate::year(docvars(data_corpus_nobel, "Date"))
load("C:/Users/kevin/Desktop/data_corpus_nobel.rdata")
temp_lengths <- stri_length(texts(data_corpus_nobel))
data_corpus_nobel <- corpus_subset(data_corpus_nobel, temp_lengths < quantile(temp_lengths, prob = .95))
nobel_corp <- corpus_trimsentences(data_corpus_nobel, min_length = 4)
nobel_lengths<-ntoken(nobel_corp, removePunct = T)
nobel_stat <- textstat_readability(nobel_corp, measure = c("Flesch", "meanSentenceLength", "meanWordSyllables"))
nobel_year <- lubridate::year(docvars(data_corpus_nobel, "Date"))
docvars(data_corpus_nobel, "Date")
docvars(data_corpus_nobel)
nobel_year <- lubridate::year(docvars(data_corpus_nobel, "year"))
nobel_year <- (docvars(data_corpus_nobel, "year"))
install.packages("randomizr", repos="http://r.declaredesign.org")
library("randomizr", lib.loc="~/R/win-library/3.4")
remove.packages("randomizr", lib="~/R/win-library/3.4")
install.packages("randomizr", repos="http://r.declaredesign.org")
install.packages("randomizr", repos = "http://r.declaredesign.org")
install.packages("randomizr", repos="http://r.declaredesign.org")
install.packages("randomizr")
library("randomizr", lib.loc="~/R/win-library/3.4")
library(DeclareDesign)
library(quanteda)
library(sophistication)
data("data_corpus_Crimson")
library(dplyr)
require(stringr)
require(data.table)
##read in
load("C:/Users/kevin/Documents/GitHub/sophistication-papers/analysis_article/output/fitted_BT_model.Rdata")
results<-predict_readability(BT_best, newdata = data_corpus_SOTU, bootstrap_n = 10, verbose = T)
data(data_corpus_sotu, package = "quanteda.corpora")
data("data_corpus_SOTU")
library(quanteda)
install.packages("quanteda")
install.packages("quanteda")
library(quanteda)
data_corpus_SOTU
results<-predict_readability(BT_best, newdata = data_corpus_SOTU, bootstrap_n = 10, verbose = T)
library(sophistication)
results<-predict_readability(BT_best, newdata = data_corpus_SOTU, bootstrap_n = 10, verbose = T)
data(data_corpus_sotu, package = "quanteda.corpora")
library(quanteda.quanteda)
library(quanteda.corpora)
install.packages("quanteda.corpora")
devtools::install_github("quanteda/quanteda.corpora")
devtools::install_github("quanteda/quanteda.corpora")
devtools::install_github("quanteda/quanteda.corpora", force = TRUE)
install.packages("digest")
library(readtext)
xx<-readtext(xx.txt)
xx<-readtext("xx.txt")
xx<-readtext(file ="xx.txt")
?readtext
xx<-readtext(file ="xx")
setwd("C:/Users/kevin/Documents/GitHub/YT_mapping_replication/")
yt_sj<-read.csv("data/youtube_sj_100.csv", header = T, stringsAsFactors = F)
yt_sj$month<-substr(yt_sj$video_publish_date, 1, 7)
yt_feminism<-read.csv("data/youtube_feminism_100.csv", header = T, stringsAsFactors = F)
yt_feminism$month<-substr(yt_feminism$video_publish_date, 1, 7)
yt_white<-read.csv("data/youtube_white_100.csv", header = T, stringsAsFactors = F)
yt_white$month<-substr(yt_white$video_publish_date, 1, 7)
ain<-read.csv(file = 'data/AIN_channel_stats.csv', header = TRUE, stringsAsFactors = F)
msm<-read.csv(file = 'data/media_yt.csv', header = TRUE, stringsAsFactors = F)
msm<-read.csv(file = 'data/media_yt.csv', header = TRUE, stringsAsFactors = F)
library(tidyverse)
yt_sj_ain<-filter(yt_sj, channel_id %in% ain$id)
yt_sj_msm<-filter(yt_sj, channel_id %in% msm$channel_id)
yt_feminism_ain<-filter(yt_feminism, channel_id %in% ain$id)
yt_feminism_msm<-filter(yt_feminism, channel_id %in% msm$channel_id)
yt_white_ain<-filter(yt_white, channel_id %in% ain$id)
yt_white_msm<-filter(yt_white, channel_id %in% msm$channel_id)
months<-yt_feminism$month
months<-unique(yt_feminism$month)
for(i in 1:length(months)){
apropos<-filter(yt_sj_ain, month == months[i])
sj_ain_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_sj_msm, month == months[i])
sj_msm_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_feminism_ain, month == months[i])
feminism_ain_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_feminism_msm, month == months[i])
feminism_msm_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_white_ain, month == months[i])
white_ain_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_white_msm, month == months[i])
white_msm_density[i]<-length(apropos$channel_title)
}
###initialize
sj_ain_density<-vector()
sj_msm_density<-vector()
feminism_msm_density<-vector()
feminism_ain_density<-vector()
white_msm_density<-vector()
white_ain_density<-vector()
##for each month
months<-unique(yt_feminism$month)
for(i in 1:length(months)){
apropos<-filter(yt_sj_ain, month == months[i])
sj_ain_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_sj_msm, month == months[i])
sj_msm_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_feminism_ain, month == months[i])
feminism_ain_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_feminism_msm, month == months[i])
feminism_msm_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_white_ain, month == months[i])
white_ain_density[i]<-length(apropos$channel_title)
apropos<-filter(yt_white_msm, month == months[i])
white_msm_density[i]<-length(apropos$channel_title)
}
alldens<-data.frame(months, feminism_msm_density, feminism_ain_density, white_msm_density, white_ain_density,
sj_msm_density, sj_ain_density)
alldens.m = melt(alldens, id.vars ="months", measure.vars = c("feminism_msm_density","feminism_ain_density", "white_msm_density","white_ain_density",
"sj_msm_density","sj_ain_density"))
library(reshape2)
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14))
alldens.m = melt(alldens, id.vars ="months", measure.vars = c("feminism_msm_density","feminism_ain_density", "white_msm_density","white_ain_density",
"sj_msm_density","sj_ain_density"))
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14))
sj_ain<-vector()
sj_msm<-vector()
feminism_msm<-vector()
feminism_ain<-vector()
white_msm<-vector()
white_ain<-vector()
for(i in 1:132){
apropos<-filter(yt_sj_ain, month == months[i])
sj_ain[i]<-length(apropos$channel_title)
apropos<-filter(yt_sj_msm, month == months[i])
sj_msm[i]<-length(apropos$channel_title)
apropos<-filter(yt_feminism_ain, month == months[i])
feminism_ain[i]<-length(apropos$channel_title)
apropos<-filter(yt_feminism_msm, month == months[i])
feminism_msm[i]<-length(apropos$channel_title)
apropos<-filter(yt_white_ain, month == months[i])
white_ain[i]<-length(apropos$channel_title)
apropos<-filter(yt_white_msm, month == months[i])
white_msm[i]<-length(apropos$channel_title)
}
alldens<-data.frame(months, feminism_msm, feminism_ain, white_msm, white_ain,
sj_msm, sj_ain)
alldens.m = melt(alldens, id.vars ="months", measure.vars = c("feminism_msm","feminism_ain", "white_msm","white_ain",
"sj_msm","sj_ain"))
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14))
alldens.m = melt(alldens, id.vars ="months", measure.vars = c("feminism_msm","white_msm","sj_msm","feminism_ain", "white_ain",
"sj_ain"))
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month: Niche Topics") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14))
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month: Niche Topics") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14), legend.title = element_blank())
pdf("figures/frequency_niche_topics.pdf", 7, 5)
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month: Niche Topics") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14), legend.title = element_blank())
dev.off()
library(tidyverse)
library(reshape2)
setwd("C:/Users/kevin/Documents/GitHub/YT_mapping_replication/")
setwd("C:/Users/kevin/Documents/GitHub/YT_mapping_replication/")
yt_economy<-read.csv("data/youtube_economy_100_US.csv", header = T, stringsAsFactors = F)
yt_economy$month<-substr(yt_economy$video_publish_date, 1, 7)
yt_news<-read.csv("data/youtube_news_100.csv", header = T, stringsAsFactors = F)
yt_economy<-read.csv("data/youtube_economy_100.csv", header = T, stringsAsFactors = F)
yt_economy$month<-substr(yt_economy$video_publish_date, 1, 7)
###news
yt_news<-read.csv("data/youtube_news_100.csv", header = T, stringsAsFactors = F)
yt_news$month<-substr(yt_news$video_publish_date, 1, 7)
###politics
yt_politics<-read.csv("data/youtube_politics_100.csv", header = T, stringsAsFactors = F)
yt_politics$month<-substr(yt_politics$video_publish_date, 1, 7)
#####now AIN // MSM channels
ain<-read.csv(file = 'data/AIN_channel_stats.csv', header = TRUE, stringsAsFactors = F)
msm<-read.csv(file = 'data/media_yt.csv', header = TRUE, stringsAsFactors = F)
####
##restrict to matches
yt_economy_ain<-filter(yt_economy, channel_id %in% ain$id)
yt_economy_msm<-filter(yt_economy, channel_id %in% msm$channel_id)
yt_news_ain<-filter(yt_news, channel_id %in% ain$id)
yt_news_msm<-filter(yt_news, channel_id %in% msm$channel_id)
yt_politics_ain<-filter(yt_politics, channel_id %in% ain$id)
yt_politics_msm<-filter(yt_politics, channel_id %in% msm$channel_id)
###initialize
economy_ain<-vector()
economy_msm<-vector()
news_msm<-vector()
news_ain<-vector()
politics_msm<-vector()
politics_ain<-vector()
##for each month
months<-unique(yt_news$month)
for(i in 1:length(months)){
apropos<-filter(yt_economy_ain, month == months[i])
economy_ain[i]<-length(apropos$channel_title)
apropos<-filter(yt_economy_msm, month == months[i])
economy_msm[i]<-length(apropos$channel_title)
apropos<-filter(yt_news_ain, month == months[i])
news_ain[i]<-length(apropos$channel_title)
apropos<-filter(yt_news_msm, month == months[i])
news_msm[i]<-length(apropos$channel_title)
apropos<-filter(yt_politics_ain, month == months[i])
politics_ain[i]<-length(apropos$channel_title)
apropos<-filter(yt_politics_msm, month == months[i])
politics_msm[i]<-length(apropos$channel_title)
}
alldens<-data.frame(months, news_msm, news_ain, politics_msm, politics_ain,
economy_msm, economy_ain)
alldens.m = melt(alldens, id.vars ="months", measure.vars = c("news_msm","politics_msm","economy_msm","news_ain", "politics_ain",
"economy_ain"))
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month: Major Topics") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14), legend.title = element_blank())
pdf("figures/frequency_major_topics.pdf", 7, 5)
ggplot(alldens.m, aes(y=value, x=months, group=variable, color = variable)) +
geom_smooth(se = F, span = .2, size = 2)+ scale_x_discrete(breaks=c("2008-01","2009-01","2010-01",
"2011-01", "2012-01", "2013-01",
"2014-01", "2015-01", "2016-01",
"2017-01", "2018-01"))+
geom_point() +
xlab(NULL) + ylab("Frequency") + theme_bw() + ggtitle("Videos in the Top 100 per Month: Major Topics") +
theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(angle = 45,
size = 14), legend.title = element_blank())
dev.off()
setwd("C:/Users/kevin/Dropbox/youtube/analysis/")
options(scipen=10000)
meta<-read.csv("data/vid_meta_msm.csv", header = T, stringsAsFactors = F)
setwd("C:/Users/kevin/Documents/GitHub/YT_mapping_replication/")
meta<-read.csv("data/vid_meta_msm.csv", header = T, stringsAsFactors = F)
meta<-filter(meta, is.na(meta$video_view_count) == F & is.na(meta$video_like_count) == F & is.na(meta$video_dislike_count) == F &
is.na(meta$video_comment_count) == F )
meta<-meta %>% distinct(video_id, .keep_all = TRUE)
meta<-filter(meta, month >= "2006-08-01" & month < "2018-11-01")
meta$month<-as.Date(paste(substr(meta$video_publish_date, 1, 7),"-01",sep=""))
meta<-filter(meta, month >= "2006-08-01" & month < "2018-11-01")
meta$video_view_count <- as.numeric(meta$video_view_count)
summary(meta$video_view_count)
meta<-filter(meta, is.na(meta$video_view_count) == F & is.na(meta$video_like_count) == F & is.na(meta$video_dislike_count) == F &
# is.na(meta$video_comment_count) == F )
####drop duplicated video_id's -- lose 2k vids here
#meta<-meta %>% distinct(video_id, .keep_all = TRUE)
##I ran this scraper midway through November 2018, so it doesn't count as a full month (lose 30k vids here)
#meta$month<-as.Date(paste(substr(meta$video_publish_date, 1, 7),"-01",sep=""))
#meta<-filter(meta, month >= "2006-08-01" & month < "2018-11-01")
summary(meta$video_view_count)
meta<-select(meta, video_view_count, video_like_count , video_comment_count , month)
##calculate video categories
table(meta$video_category)/length(meta$video_id)
sum_likes<-aggregate(video_like_count ~ dates, data = meta,
sum)
pdf(file = "sum_likes_msm.pdf", width = 8, height = 5)
qplot( sum_likes$dates, sum_likes$video_like_count, geom = 'smooth' )+ theme(
plot.background = element_rect(fill = "white"),
panel.background = element_rect(fill = "white"),
axis.line.x = element_line(color = "grey")
) + ylab("Sum Likes per Day") + ylim(0,NA) +ggtitle("MSM Likes Per Day ")
dev.off()
mean_views<-aggregate(video_view_count ~ dates, data = meta,
mean)
sum_views<-aggregate(video_view_count ~ dates, data = meta,
sum)
pdf(file = "sum_views_msm.pdf", width = 8, height = 5)
#plot(mean_views$dates , (mean_views$video_view_count), xlab = "Date", ylab = "Mean Views", col = "red")
qplot( sum_views$dates, sum_views$video_view_count, geom = 'smooth' ) + theme(
plot.background = element_rect(fill = "white"),
panel.background = element_rect(fill = "white"),
axis.line.x = element_line(color = "grey")
) + xlab("Date") + ylab("Sum Views per Day") + ylim(0,NA) +ggtitle("MSM Views Per Day ")
dev.off()
#scatter.smooth(as.Date(mean_views$dates), (mean_views$video_view_count), col = "red", span = .1)
vids_ain[148]
vids_msm<-table(meta$month)
vids_msm_short<-vids_msm
#vids_msm_short<-vids_msm[9:156]
vids_msm_short[vids_msm_short>15000]<-15000
pdf(file = "vids_msm_short.pdf", width = 8, height = 5)
plot(vids_msm_short, ylab = "Videos Uploaded per Month", main = "Videos by MSM")
meta<-filter(meta, is.na(meta$video_view_count) == F & is.na(meta$video_like_count) == F & is.na(meta$video_dislike_count) == F &
is.na(meta$video_comment_count) == F )
meta<-select(meta, video_view_count, video_like_count , video_comment_count , month)
write.csv(file ="data/vid_meta_msm_preprocessed.csv", x = meta)
meta<-read.csv("data/vid_meta_msm.csv", header = T, stringsAsFactors = F)
meta<-read.csv("data/vid_meta_msm_preprocessed.csv", header = T, stringsAsFactors = F)
table(meta$video_category)/length(meta$video_id)
sum_likes<-aggregate(video_like_count ~ month, data = meta,
sum)
qplot( sum_likes$dates, sum_likes$video_like_count, geom = 'smooth' )+ theme(
plot.background = element_rect(fill = "white"),
panel.background = element_rect(fill = "white"),
axis.line.x = element_line(color = "grey")
) + ylab("Sum Likes per Month") + ylim(0,NA) +ggtitle("MSM Likes Per Month ")
qplot( sum_likes$month, sum_likes$video_like_count, geom = 'smooth' )+ theme(
plot.background = element_rect(fill = "white"),
panel.background = element_rect(fill = "white"),
axis.line.x = element_line(color = "grey")
) + ylab("Sum Likes per Month") + ylim(0,NA) +ggtitle("MSM Likes Per Month ")
View(sum_likes)