What can CATS bring to researchers who want to analyze tweets?
Michael Gauthier
Université Lyon
Twitter's API (Application Programming Interface) makes it possible for anyone to collect tweets for various purposes, and the increasing number of studies based on tweets confirms the usefulness of such corpora. However, the programing knowledge required to carry this out is a problem for a lot of people who do not have those skills, or the resources to work with computer programers, and this is especially true in social sciences. CATS (Collection and Analysis of Tweets made Simple) is a web interface developed by researchers from the universities of Lyon 1 and 2, and its purpose is to enable everyone to generate corpora of tweets with specific demands (keywords, geolocation, language etc...), even for people without any programing background. Not only does CATS allow people to collect tweets, it also enables one to analyze them thanks to various tools (vocabulary exploration, topic modeling, event detection, word clouds).
The goal of this presentation will be to introduce CATS, and what it can do thanks to a demonstration of the tool, and some practical examples based on previously collected corpora. However, the interface also has certain limitations, that I will be happy to discuss as well.