Analyzing #lastesis feminist movement in twitter using topic models

Abstract

Nowadays, social networks have created a massive mean of communication, that was unthinkable many years ago. Informal communication, blogging, and online discussions have transformed the Web into a huge repository of remarks on numerous themes, producing a potential wellspring of data for various areas. In this paper we analyze, using Topic Models, a recent widespread feminist movement. Las Tesis is a feminist collective that initiated a protest against sexual abuse, and that was replicated in more than dozen different countries in matter of days. We use LDA and BTM to detect automatically the topics in over 627643 tweets that were gathered from the 25th November until the 5th January. The resulting topics obtained, from tweets in Spanish and English, show that these algorithms are able to capture the real-world events that occurred in Chile and Turkey.