install.packages("rtweet")
install.packages("revgeo")
install.packages("tm")
install.packages("tidytext")
install.packages("igraph")
install.packages("dplyr")
install.packages("ggplot2")
install.packages("tidyverse")
install.packages("stringr")
install.packages("scales")
install.packages("furrr")
install.packages("wydir")
install.packages("ggraph")


library(rtweet)
library(revgeo)
library(tm)
library(tidytext)
library(igraph)
library(dplyr)
library(ggplot2)
library(tidyverse)
library(stringr)
library(knitr)
library(scales)


library(rtweet)
twitter_token <- create_token(
  app = "ConspApp",
  consumer_key = "3XTcYTCmb5JcRmsvctfBupaBF",
  consumer_secret = "QRaEJD2Lrb1ymeWtvVctr4XG5FJRjlaAZO3dnH3zB0oDmyKwo9",
  access_token="800311264824041472-x0RKc6Ep1F7JeEMozKsqP4zj45JG398",
  access_secret = "hnZTXa2d2jpjPabe1J4xRAnGcJEamRJnf5KzkiuR5S17a")

## search for 18000 tweets using the metoo hashtag
rt <- search_tweets(
  "#MeToo", n = 18000, include_rts = FALSE
)

## search for 250,000 tweets containing the word data
rt <- search_tweets(
  "data", n = 250000, retryonratelimit = TRUE
)

## search for 10,000 tweets sent from the US
rt <- search_tweets(
  "MeToo", geocode = lookup_coords("usa"), n = 1000
)


## create lat/lng variables using all available tweet and profile geo-location data
rt <- lat_lng(rt)

## plot state boundaries
par(mar = c(0, 0, 0, 0))
maps::map("state", lwd = .25)

## plot lat and lng points onto state map
with(rt, points(lng, lat, pch = 20, cex = .75, col = rgb(0, .3, .7, .75)))

## random sample for 30 seconds (default)
rt <- stream_tweets("")

## stream tweets for a week (60 secs x 60 mins * 24 hours *  7 days)
Rt<-stream_tweets(
  "",timeout = 60 * 60 * 24 * 7,)

#Get available trends
trends <- trends_available()

#See all trends of a specific city
Milan<-get_trends(718345)


## get user IDs of accounts followed by Matteo Salvini
Matteo_friends <- get_friends("@matteosalvinimi")

## lookup data on those accounts
Matteo_friends_data <- lookup_users(Matteo_friends$user_id)

## get user IDs of accounts following Matteo Salvini
Matteo_followers <- get_followers("@matteosalvinimi", n = 75000)

## lookup data on those accounts
Matteo_followers_data <- lookup_users(Matteo_followers$user_id)

tmls <- get_timelines(c("@GiorgiaMeloni", "@matteosalvinimi", "@luigidimaio","@nzingaretti","@emmabonino","@lauraboldrini","@matteorenzi"), 
                      n = 320)

#Frequency by time
Data_series<-ts_data(rt, by = "mins", trim = 0L, tz = "UTC")

#Plot time series
ts_plot(rt,by="mins")



##For writing your data in csv format
attach (rt)
#OR
attach (tmls)

tweetsc<-data.frame(user_id,screen_name,created_at,text,favorite_count,retweet_count,lang,location,followers_count,friends_count,lat,lng,name,listed_count,statuses_count)
write.csv2(tweetsc,"filename.csv")

#############################################################################################

library(RedditExtractoR)
  
#Find subreddits
find_subreddits(keywords)


## Look for links containing the word conspiracy
  links <- find_thread_urls(
    keywords = "conspiracy",
    sort_by = "top",
    subreddit = NA,
    period = "month"
  )
  
  #get the content
  content <- get_thread_content(links)
  comments<-content$comments
  threads<-content$threads
  View(comments)
  View(threads)
  
#Find users contents
  get_user_content(users)


  
## Create a csv dataset
  write.cvs2(threads,"filename.csv")

#=======================================================================================

  #STM
  library(stm)
  library(readxl)
  processed <- textProcessor(data$text, metadata = data,
                             lowercase = TRUE, #*
                             removestopwords = TRUE, #*
                             removenumbers = TRUE, #*
                             removepunctuation = TRUE, #*
                             stem = TRUE, #*
                             wordLengths = c(3,Inf), #*
                             sparselevel = 1, #*
                             language = "en", #*
                             verbose = TRUE, #*
                             onlycharacter = TRUE, # not def
                             striphtml = FALSE, #*
                             customstopwords = NULL, #*
                             v1 = FALSE) #*
  out<-prepDocuments(processed$documents, processed$vocab, processed$meta)
  meta<-out$meta
  vocab<-out$vocab
  docs<-out$documents
  findingk <- searchK(out$documents, out$vocab, K = c(2:30), data = out$meta, verbose=FALSE)
  First_STM <- stm(documents = out$documents, vocab = out$vocab,
                   K = 10, prevalence =~ 1,
                   max.em.its = 75, data = out$meta,
                   init.type = "Spectral", verbose = TRUE) ### 
  
  plot(First_STM)
  topics<-labelTopics(First_STM,n=15) #see the topic
  
  
  predict_topics<-estimateEffect(formula = 1:6 ~ power, stmobj = First_STM, metadata
                                 = out$meta, uncertainty = "Global")
  summary(predict_topics)