In this demo paper, we present NEED4Tweet, a Twitterbot for named entity extraction (NEE) and disambiguation (NED) for Tweets. The straightforward application of state-of-the-art extraction and disambiguation approaches on informal text widely used in Tweets, typically results in significantly degraded performance due to the lack of formal structure; the lack of sufficient context required; and the seldom entities involved. In this paper, we introduce a novel framework that copes with the introduced challenges. We rely on contextual and semantic features more than syntactic features which are less informative. We believe that disambiguation can help to improve the extraction process. This mimics the way humans understand language.
|Title of host publication
|Proceedings of the System Demonstrations of The 53rd Annual Meeting of the Association for Computational Linguistics (ACL 2015), Beijing, China
|Place of Publication
|The Association for Computer Linguistics
|Published - 1 Jul 2015