Blog 4

Text As Data.

library(jsonlite)
library(dplyr)
library(sentimentr)
library(tidyverse)
library(quanteda)
library(readr)
library(tm)

So I ended up switching from rtweet to Academic Twitter API and ended up paying for premium twitter access since my request rates went over the limit. Below I read in the RDS file and convert tweets to ASCII to avoid character issues and emojis. I also remove html links, stopwords, punctuation, tabs, and the @ before usernames.

tweets <- readRDS("~/networks-blog/boliviatweetsfile")
tweets <- iconv(tweets$text, to = "ASCII", sub = " ")
tweets <- gsub("(RT|via)((?:\\b\\W*@\\w+)+)", "", tweets)  # Remove the "RT" (retweet) and usernames 
tweets = gsub("http.+ |http.+$", " ", tweets)  # Remove html links
tweets = gsub("http[[:alnum:]]*", "", tweets)
tweets = gsub("[[:punct:]]", " ", tweets)  # Remove punctuation
tweets = gsub("[ |\t]{2,}", " ", tweets)  # Remove tabs
tweets = gsub("^ ", "", tweets)  # Leading blanks
tweets = gsub(" $", "", tweets)  # Lagging blanks
tweets = gsub(" +", " ", tweets) # General spaces 
tweets = tolower(tweets)
tweets = unique(tweets)
glimpse(tweets)

 chr [1:40843] "1rudecruzan dredadonx maadynyc bolivia staged a mass strike after america staged a coup and removed their democ"| __truncated__ ...

Creating the corpus

corpus <- Corpus(VectorSource(tweets))
corpus <- tm_map(corpus, removeWords, stopwords("english"))  
corpus <- tm_map(corpus, removeNumbers)

corpus <- tm_map(corpus, stemDocument)
corpus = tm_map(corpus, removeWords, c("amp", "will", "get", "can", "like", "say", "know"))

Wordcloud of words mentioned at least 1,000 times. It’s clear that bolivia and coup are prominent so I want to remove them and see what the wordcloud looks like after.

library(wordcloud)
set.seed(100)
wordcloud(corpus, min.freq = 1000)

corpus2 <- Corpus(VectorSource(tweets))
corpus2 <- tm_map(corpus, removeWords, stopwords("english"))  
corpus2 <- tm_map(corpus, removeNumbers)

corpus2 <- tm_map(corpus, stemDocument)
corpus2 = tm_map(corpus, removeWords, c("bolivia", "coup", "amp", "will", "get", "can", "like", "say", "know"))

This gives a clearer picture on what’s going on. We see some words such as support,elect, peopl so I’m wondering if these are connected in a positive or negative way to the political crisis.

library(wordcloud)
set.seed(100)
wordcloud(corpus2, min.freq = 1000)