TwittDict: Extracting Social Oriented Keyphrase Semantics from Twitter
Suppawong Tuarob, Wanghuan Chu, Dong Chen, Conrad S. Tucker · 2015
Social media not only carries information that is up-to-date, but also bears the wisdom of the crowd.In social media, new words are developed everyday, including slangs, combinations of existing terms, entity names, etc.These terms are initially used in small communities, which can later grow popular and become new standards.The ability to early recognize the existence and understand the meanings of these terms can prove to be crucial, especially to emergence detection applications.We present an ongoing research work that investigates the use of topical analysis to extract semantic of terms in social media.In particular, the proposed method extracts semantically related words associated with a target word from a corpus of tweets.We provide preliminary, anecdotal results comprising the semantic extraction of five different keywords.