Paper: Umigon: sentiment analysis for tweets based on terms lists and heuristics

ACL ID S13-2068
Title Umigon: sentiment analysis for tweets based on terms lists and heuristics
Venue Joint Conference on Lexical and Computational Semantics
Session
Year 2013
Authors

Umigon is developed since December 2012 as a web application providing a service of sentiment detection in tweets. It has been designed to be fast and scalable. Umigon also provides indications for additional semantic features present in the tweets, such as time indications or markers of subjectivity. Umigon is in continuous development, it can be tried freely at www.umigon.com. Its code is open sourced at: https://github.com/seinecle/Umigon 1. General principle of operation Umigon belongs to the family of lexicon based sentiment classifiers (Davidov et al. 2010, Kouloumpis et al. 2011). It is specifically designed to detect sentiment (positive, negative or neutral) in tweets. The ?sentiment detection engine? of Umigon consists of 4 main parts, which are detailed below: ...