Twitter Plot

Twitter Plot

setwd("/Users/subasishdas1/Desktop/tweets/tweets2/02")
library(readxl)
dat= read_excel("trbam.xlsx")
dat=head(dat, 200)

library(tidyverse)    

clean_tweets <- function(x) {
  x %>%
    str_remove_all(" ?(f|ht)(tp)(s?)(://)(.*)[.|/](.*)") %>%
    str_replace_all("&amp;", "and") %>%
    str_remove_all("[[:punct:]]") %>%
    str_remove_all("^RT:? ") %>%
    str_remove_all("@[[:alnum:]]+") %>%
    str_remove_all("#[[:alnum:]]+") %>%
    str_remove_all("pictwittercom") %>%
    str_replace_all("\\\n", " ") %>%
    str_to_lower() %>%
    str_trim("both")
}

dat$tweet1= dat$tweet %>% clean_tweets
dat$hashtags1= dat$hashtags %>% clean_tweets
library(qdapRegex)
dat$mentions2= dat$mentions %>% clean_tweets
dat$mentions3=stringr::str_extract_all(dat$mentions2, regex("(?<=screenname\\s)[:alpha:]*\\b"))


names(dat)

##  [1] "id"              "conversation_id" "created_at"      "date"           
##  [5] "time"            "timezone"        "user_id"         "username"       
##  [9] "name"            "place"           "tweet"           "language"       
## [13] "mentions"        "urls"            "photos"          "replies_count"  
## [17] "retweets_count"  "likes_count"     "hashtags"        "cashtags"       
## [21] "link"            "retweet"         "quote_url"       "video"          
## [25] "thumbnail"       "near"            "geo"             "source"         
## [29] "user_rt_id"      "user_rt"         "retweet_id"      "reply_to"       
## [33] "retweet_date"    "translate"       "trans_src"       "trans_dest"     
## [37] "tweet1"          "hashtags1"       "mentions2"       "mentions3"

library(igraph)
library(hrbrthemes)
library(ggraph)
library(tidyverse)

filter(dat, retweets_count > 0) %>% 
  select(username, mentions3) %>%
  unnest(mentions3) %>% 
  filter(!is.na(mentions3)) %>% 
  graph_from_data_frame() -> rt_g

V(rt_g)$node_label <- unname(ifelse(degree(rt_g)[V(rt_g)] > 2, names(V(rt_g)), "")) 
V(rt_g)$node_size <- unname(ifelse(degree(rt_g)[V(rt_g)] > 2, degree(rt_g), 0)) 


ggraph(rt_g, layout = 'linear', circular = TRUE) + 
  geom_edge_arc(edge_width=0.125, aes(alpha=..index..)) +
  geom_node_label(aes(label=node_label, size=node_size),
                  label.size=0, fill="#ffffff66", segment.colour="springgreen",
                  color="slateblue", repel=TRUE, fontface="bold") +
  coord_fixed() +theme_bw()+
  theme(axis.title=element_blank(),
        axis.text=element_blank(),
        axis.ticks=element_blank())+
  scale_size_area(trans="sqrt") +
  theme(legend.position="none")

dat$mentions4= dat$mentions3 %>% clean_tweets
names(dat)

##  [1] "id"              "conversation_id" "created_at"      "date"           
##  [5] "time"            "timezone"        "user_id"         "username"       
##  [9] "name"            "place"           "tweet"           "language"       
## [13] "mentions"        "urls"            "photos"          "replies_count"  
## [17] "retweets_count"  "likes_count"     "hashtags"        "cashtags"       
## [21] "link"            "retweet"         "quote_url"       "video"          
## [25] "thumbnail"       "near"            "geo"             "source"         
## [29] "user_rt_id"      "user_rt"         "retweet_id"      "reply_to"       
## [33] "retweet_date"    "translate"       "trans_src"       "trans_dest"     
## [37] "tweet1"          "hashtags1"       "mentions2"       "mentions3"      
## [41] "mentions4"

dat1= dat[, c(8, 41)]
net <- dat1 %>% 
  group_by(mentions4) %>% 
  mutate(count = n()) %>% 
  ungroup() %>% 
  filter(count > 1)
library(tidygraph)

net1= subset(net, mentions4!="character0")
glimpse(net1)

## Rows: 38
## Columns: 3
## $ username  <chr> "gaellebris", "sctlcenter", "sctlcenter", "kittelson", "con…
## $ mentions4 <chr> "cnasemtrb wspusa", "nasemtrb", "c uwengineering uwmst nase…
## $ count     <int> 2, 28, 2, 28, 28, 2, 28, 2, 28, 28, 2, 28, 2, 28, 28, 2, 28…

g2 <- net1 %>% 
  select(username, mentions4) %>%  # drop the count column
  as_tbl_graph(directed = F) %>%  # make undirected
  activate(nodes) %>% 
  mutate(centrality = centrality_betweenness())
g2

## # A tbl_graph: 24 nodes and 38 edges
## #
## # An undirected multigraph with 1 component
## #
## # Node Data: 24 x 2 (active)
##   name            centrality
##   <chr>                <dbl>
## 1 gaellebris            22  
## 2 sctlcenter            22.0
## 3 kittelson              0  
## 4 congestionfixes        0  
## 5 nasemtrb             248. 
## 6 uwmst                 43.0
## # … with 18 more rows
## #
## # Edge Data: 38 x 2
##    from    to
##   <int> <int>
## 1     1    21
## 2     2     5
## 3     2    22
## # … with 35 more rows

ggraph(g2) +
  geom_edge_link() +
  geom_node_point(aes(size = centrality, colour = centrality)) +
  theme_graph()