**DataScience+**, and kindly contributed to R-bloggers)

The following R-script is for download the data automatically from twitter using the package `SchedulerR`

. As first step we download the data using the `OAuth protocol`

and store the data using as a name the date from the download. As second we used the package `SchedulerR`

for to specify how long the script runs. This package has the advantage that we can set the time and also how long will repeat either daily, weekly or monthly.

Load the packages:

library(twitteR) library(httr) library(SchedulerR)

Also we choose the directory where the RData would be saved:

setwd("C:/Users/Documents/schedulerR")

OAuth protocol:

api_key <- "xxxxxxxxxxxxxxxxxxxxxx" api_secret <- "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" access_token <- "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" access_token_secret <- "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" setup_twitter_oauth(api_key,api_secret,access_token,access_token_secret)

Extract the tweets using usertimeline function from the Twitter R package:

clinton_tweets <- userTimeline("HillaryClinton", n = 3200) tweetsc.df <- twListToDF(clinton_tweets) dim(tweetsc.df)

Create a variable called date and become to character:

date<-Sys.Date() date<-as.character(date) name<-paste(date,".RData") #Finally we save the Rdata using as name the date from the download save(tweetsc.df, file =name)

The script before is for download the data from twitter.

The next step is used the package SchedulerR, the script for this is the following:

taskscheduler_create(taskname = "taskclinton", rscript = clintontweets, schedule = "DAILY", starttime = "11:30", startdate = format(Sys.Date(), "%d/%m/%Y"))

Also we can used the add in interface on Rstudio add in tasks. Feel free to post comment if you have any question or suggestion about this post.

## Examples of Analysis

We can implement many analysis from text data.

Find words common with a minimun frequency of 45:

term.freq <- rowSums(as.matrix(tdm)) term.freq = 45) #we saved the data in a data frame df <- data.frame(term = names(term.freq), freq = term.freq) save(df,file="tqfclinton50.RData") #plot with ggplot2 library(ggplot2) ggplot(df, aes(x = term, y = freq)) + geom_bar(stat = "identity") + xlab("Terms") + ylab("Count") + coord_flip()

Hierarchical cluster:

# remove the parse terms tdm2 <- removeSparseTerms(tdm, sparse = 0.95) m2 <- as.matrix(tdm2) # hierarchical cluster m <- as.matrix(m2) d<- dist(m2) #we used the cosine similaruty as distance because is the best option for text data distcos<-dissimilarity(x=m2,method='cosine') groups <- hclust(distcos,method="complete") #plot dendograma plot(groups, hang=-1) rect.hclust(groups,5)

### K means clustering

We combine the Kmeans clustering with Principal components analysis for obtain a better result. As a first we used the elbow method for choose the “K”

We choose K=5 from the plot. Now we proceed to make the K means clustering.

pc <- princomp(tdm) plot(pc) data<-tdm wss <- (nrow(data)-1)*sum(apply(data,2,var)) for (i in 2:15) wss[i] <- sum(kmeans(data, centers=i)$withinss) plot(1:15, wss, type="b", xlab="Number of Clusters", ylab="Within groups sum of squares") m3 <- t(tdm2) # transpse from term document matriz set.seed(1222) # we choose a seed. k <- 5# number of clusters kmeansResult <- kmeans(m3, k) km<-round(kmeansResult$centers, digits = 3) #we saved in a csv file. write.csv(km,file="kmeansclinton.csv")

The following loop is for visualize the clusters.

for (i in 1:k) { cat(paste("cluster ", i, ": ", sep = "")) s <- sort(kmeansResult$centers[i, ], decreasing = T) cat(names(s)[1:5], "\n") } write.csv(s,file="clusterkm.csv")

From the results we can said from each cluster they are the topics most spoken from the tweets.

Related Post

- An Introduction to Time Series with JSON Data
- Get Your Data into R: Import Data from SPSS, Stata, SAS, CSV or TXT
- Exporting Data from R to TXT, CSV, SPSS or Stata
- First Things to Do After You Import the Data into R

**leave a comment**for the author, please follow the link and comment on their blog:

**DataScience+**.

R-bloggers.com offers

**daily e-mail updates**about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more...