This project is an attempt to create a system capable of detecting trends, events, and conversation on Twitter Public Stream.
Events are detected using detecting entity and enriched with NLP tools. A semantic search engine indexes tweets using online LDA model. A detected event can be used to query the model as to find more similar tweets and enrich/confirm an event.
--Inspired by statuscalendar.com
More documentation coming up. Let me know if anybody is interested in this subject.