Twitter Profile Analysis

This repository contains files for a social listening prototype that summarizes the network engagement and hashtag usage of a Twitter user's most recent social media messages.

The underlying processing algorithm uses regular expressions to isolate words that have been tagged with the "@" or "#" symbols and tabulates occurrence frequencies based on their appearance in the n-sized tweet corpus (in the example below, n=60).

Setup

Prerequisites

• R version 3.3.3+
• Latest version of RStudio

Installation

Open RStudio and enter the following into the interactive command line to install the twitteR package.

# Install the 'twitteR' package
install.packages("twitteR")

You will need a Twitter developer account to complete the next step. If you do not yet have one, follow these instructions before you proceed.

Open twitterConfig.R and enter your authentication credentials to access the Twitter rest API.

# Filename: twitterConfig.R
# Created: April 21, 2016 
# Description: sourcefile for scripts requiring the 'twitteR' package

library("twitteR")

consumer_key <- 'XXXXXXXX'
consumer_secret <- 'XXXXXXXX'
access_token <- 'XXXXXXXX'
access_secret <- 'XXXXXXXX'
setup_twitter_oauth(consumer_key,
                    consumer_secret,
                    access_token,
                    access_secret)

Getting Started

Now that everything is installed, open the file R Markdown file twitterReport.Rmd. Enter the name of the account for which you wish to generate a report. You may also specify the number of tweets for the algorithm to retrieve and parse by changing the numerical value of "n".

# Setup
user <- "QueriumCorp"
n <- 60

Click "Knit HTML" to create an html document of the user's most recent n tweets.

The final output will show the name and description of the target account along with 3 boxes containing the most frequently used hashtags, the users who were either retweeted or replied to, and the descriptions of the associated accounts ordered alphabetically.