This tutorial demonstrates how to use R to collect tweets and apply a (very) naive algorithm to estimate their emotional sentiment. Despite the simplicity of our algorithm and the very small size of our sample of tweets, we are nevertheless able to find an interesting result which compares well to ACSI's respected study of customer satisfaction.

This tutorial was originally presented as a first-time introduction to R for the savvy audience of the Boston Predictive Analytics Meetup Group. As such, its primary focus is to highlight R as a tool to get data easily and synthesize results quickly.

Presentation slides are available from my blog (http://jeffreybreen.wordpress.com/2011/07/04/twitter-text-mining-r-slides/) and the complete code is available on github (https://github.com/jeffreybreen/twitter-sentiment-analysis-tutorial-201107).

This work is also featured in Elsevier's forthcoming book Practical Text Mining and Statistical Analysis for Non-structured Text Data Applications by Gary Miner et al. (http://ow.ly/63NWl).