Exploiting Semantic Role Labeling, WordNet and Wikipedia for Coreference Resolution

Proceedings of the Human Language Technology Conference of the North American Chapter of the ACL, pages 192–199, New York, June 2006. c⃝2006 Association for Computational Linguistics Exploiting Semantic Role Labeling, WordNet and Wikipedia for Coreference Resolution Simone Paolo Ponzetto and Michael Strube EML Research gGmbH Schloss-Wolfsbrunnenweg 33 69118 Heidelberg, Germany http://www.eml-research.de/nlp Abstract In this paper we present an extension of a machine learning based coreference res- olution system which uses features in- duced from different semantic knowledge sources. These features represent knowl- edge mined from WordNet and Wikipedia, as well as information about semantic role labels. We show that semantic features in- deed improve the performance on differ- ent referring expression types such as pro- nouns and common nouns. 1 Introduction The last years have seen a boost of work devoted to the development of machine learning based coref- erence resolution systems (Soon et al., 2001; Ng & Cardie, 2002; Yang et al., 2003; Luo et al., 2004, inter alia). While machine learning has proved to yield performance rates fully competitive with rule based systems, current coreference resolution sys- tems are mostly relying on rather shallow features, such as the distance between the coreferent expres- sions, string matching, and linguistic form. How- ever, the literature emphasizes since the very begin- ning the relevance of world knowledge and infer- ence for coreference resolution (Charniak, 1973). This paper explores whether coreference resolu- tion can benefit from semantic knowledge sources. More specifically, whether a machine learning based approach to coreference resolution can be improved and which phenomena are affected by such infor- mation. We investigate the use of the WordNet and Wikipedia taxonomies for extracting semantic simi- larity and relatedness measures, as well as semantic parsing information in terms of semantic role label- ing (Gildea & Jurafsky, 2002, SRL henceforth). We believe that the lack of semantics in the cur- rent systems leads to a performance bottleneck. In order to correctly identify the discourse entities which are referred to in a text, it seems essential to reason over the lexical semantic relations, as well as the event representations embedded in the text. As an example, consider a fragment from the Automatic Content Extraction (ACE) 2003 data. (1) But frequent visitors say that given the sheer weight of the country’s totalitarian ideology and generations of mass indoctrination, changing this country’s course will be something akin to turning a huge ship at sea. Opening North Korea up, even modestly, and exposing people to the idea that Westerners – and South Koreans – are not devils, alone represents an extraordinary change. [...] as his people begin to get a clearer idea of the deprivation they have suffered, especially relative to their neighbors. “This is a society that has been focused most of all on stability, [...]”. In order to correctly resolve the anaphoric expres- sions highlighted in bold, it seems that some kind of lexical semantic and encyclopedic knowledge is required. This includes that North Korea is a coun- try, that countries consist of people and …

Content truncated to 500 words