Twitter logo


Sentiment analysis for tweets, and more.

tweets analyzed in Umigon
since last reboot:


Application created by Clement Levallois. Can process millions of tweets in minutes. Download the (not too technical) paper describing how Umigon works.
Code available in open source on Github.

The detection of extra semantic features has been switched off (promotional tweets, markers of subjectivity, etc.) Please let me know at if you'd like to switch it on again.
New! Umigon is now available through an API. Process millions of tweets automatically: Go to the API.

If you use this application, please reference it in your work:
Levallois, Clement. "Umigon: sentiment analysis on Tweets based on terms lists and heuristics". Proceedings of the 7th International Workshop on Semantic Evaluation (SemEval 2013), June 2013, Atlanta, Georgia.

Recommended references to explore the possibilities of text mining:
Data science for business Data smart Python 3 and NLTK 3 cookbook Taming text