-
Notifications
You must be signed in to change notification settings - Fork 0
/
Twitter.R
145 lines (112 loc) · 3.71 KB
/
Twitter.R
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
##############################################################
############### Twitter Data Analysis ########################
##############################################################
rm(list = ls())
silent(gc())
load_lb <- function()
{
suppressPackageStartupMessages(library(doMC))
registerDoMC(cores = 8)
suppressPackageStartupMessages(library(readxl))
suppressPackageStartupMessages(library(tidyr))
suppressPackageStartupMessages(library(dplyr))
suppressPackageStartupMessages(library(caret))
suppressPackageStartupMessages(require(Matrix))
suppressPackageStartupMessages(require(ggplot2))
suppressPackageStartupMessages(require(data.table))
suppressPackageStartupMessages(require(treemap))
suppressPackageStartupMessages(require(highcharter))
}
load_lb()
df <- read_excel("E:\\Study\\R Projects\\Common files\\gender-classifier-DFE-791531.xlsx")
head(df)
glimpse(df)
# 20,050 X 26
## Column names changed
colnames(df)[1:5] <- c("Unit_ID", "Golden", "Unit_state", "Trusted_jud", "Last_jud_at")
names(df)
## Finding NULL
sort(sapply(df, function(x) sum(is.na(x))))
df %>%
mutate(profile_yn_gold = NULL,
gender_gold = NULL,
tweet_coord = NULL) %>%
filter(!is.na(gender))-> df
glimpse(df)
head(df)
# no duplicate Unit_ID
length(unique(df$Unit_ID))
fct = unique(df$Golden)
df$Golden = as.integer(factor(df$Golden, labels = fct))
## Check for Golden proportion
df %>%
group_by(Golden) %>%
summarise(cnt = n()) %>%
ggplot(aes(as.factor(Golden), cnt)) +
geom_bar(stat = "identity") +
geom_text(aes(label = cnt), vjust = -0.2)+
labs(title = "Golden distribution", x = "status", y = "Count")
## Very few people in Golden status
unique(df$Unit_state)
df %>%
group_by(Unit_state) %>%
summarise(cnt = n()) %>%
ggplot(aes(Unit_state, cnt)) +
geom_bar(stat = "identity") +
geom_text(aes(label = cnt), vjust = -0.2)+
labs(title = "Unit state distribution", x = "status", y = "Count")
## redundant column
unique(df$gender)
df %>%
group_by(gender) %>%
summarise(cnt = n()) %>%
ggplot(aes(reorder(gender,cnt), cnt, fill = cnt)) +
geom_bar(stat = "identity", show.legend = FALSE) +
geom_text(aes(label = cnt), hjust = -0.2)+
labs(title = "Gender distribution", x = "Gender", y = "Count") +
coord_flip()
glimpse(df)
df %>%
mutate(year = year(created),
Month = month(created)) -> df
df %>%
group_by(year) %>%
summarise(cnt = n()) %>%
arrange(-cnt) %>%
ggplot(aes(reorder(year,cnt), cnt, fill = cnt)) +
geom_bar(stat = "identity", show.legend = FALSE)+
geom_text(aes(label = cnt), hjust = -0.1) +
coord_flip()
df %>%
group_by(Month) %>%
summarise(cnt = n()) %>%
arrange(-cnt) %>%
ggplot(aes(reorder(Month,cnt), cnt, fill = cnt)) +
geom_bar(stat = "identity", show.legend = FALSE)+
geom_text(aes(label = cnt), hjust = -0.1) +
coord_flip()
df %>%
mutate(created_dt = as.Date(as.POSIXct(created))) -> df
glimpse(df)
df %>%
filter(!is.na(user_timezone)) %>%
filter(user_timezone %in% c("Pacific Time (US & Canada)")) %>%
group_by(created_dt,user_timezone) %>%
summarise(cnt = n()) %>%
arrange(-cnt) %>%
ggplot(aes(created_dt,cnt))+
geom_line()+
facet_wrap(~user_timezone)
## Tweets by user timezone
df %>%
filter(!is.na(user_timezone)) %>%
group_by(user_timezone) %>%
summarise(cnt = n()) %>%
arrange(-cnt) %>%
top_n(10) %>%
ggplot(aes(reorder(user_timezone,cnt), cnt, fill = cnt)) +
geom_bar(stat = "identity", show.legend = FALSE)+
geom_text(aes(label = cnt), hjust = -0.1) +
coord_flip()
df_t <- as_tibble(df$text)
head(df_t)