Cross-genre Event Extraction with Knowledge Enrichment

Proceedings of NAACL-HLT 2016, pages 1158–1162, San Diego, California, June 12-17, 2016. c⃝2016 Association for Computational Linguistics Cross-genre Event Extraction with Knowledge Enrichment Hao Li and Heng Ji Computer Science Department Rensselaer Polytechnic Institute jih@rpi.edu Abstract The goal of Event extraction is to extract struc- tured information of events that are of interest from unstructured documents. Existing event extractors for social media suffer from two major problems: lack of context and informal nature. In this paper, instead of conducting event extraction solely on each social media message, we incorporate cross-genre knowl- edge to boost the event extractor performance. Experiment results demonstrate that without any additional annotations, our proposed ap- proach is able to provide 15% absolute F-score improvement over the state-of-the-art. 1 Introduction The rapid development of social media and social networks since 2000s has made it an important chan- nel of information dissemination. Because of its real-time nature, social media can be used as a sen- sor to gather up-to-date information about the state of the world. Effective automatic detection and ex- traction of events from the media will be an ex- tremely important contribution. Recently there has been increasing interests in event extraction from so- cial media (Yang et al., 1998; Kleinberg, 2003; He et al., 2007; Weng and Lee, 2011; Benson et al., 2011; Ritter et al., 2012). Identifying and extracting events in social media is more challenging than traditional event extraction due to two major reasons: (1). Lack of Context: compared with traditional genres (e.g., new articles), social media context is usually short and incomplete (e.g., each tweet has a length limitation of 140 char- acters). Lacking of context, a single tweet itself usu- ally cannot provide a complete picture of the corre- sponding events. For example, for the tweet “Pray for Mali - the situation is coming to light, and it isn’t pretty.”, an event extraction/discovery system (e.g. (Ji and Grishman, 2008)) fails to discover that it is about the same war event in Mali as mentioned in the news article “State military forces on Friday re- took a key town in northern Mali after intense fight- ing that included help from French military forces, a defense ministry spokesman said.” (2). Informal Nature: social media messages are written in an in- formal style, which causes the poor performance of event extractors designed for formal genres. For ex- ample, the tweet “#AaronSwartz, Dead @ 26, #Car- menOrtiz “pushed him to exhaustion” don’t let her get away with this! #scandal.” includes an “Die” event with Aaron Swartz as the Victim. However, the person name “AaronSwartz” appears in the hash- tag “#AaronSwartz” and “Dead at 26” is written as “Dead @ 26”. Existing supervised name taggers and event extractors fail to identify the same “Die” event mentioned in the news article “Internet ac- tivist Aaron Swartz dead at 26”. Based on the intuition that news articles contain more detailed and formal information than tweet messages, we apply an unsupervised knowledge en- richment algorithm to link each tweet to …

Content truncated to 500 words