Anaphora Resolution for Biomedical Literature by Exploiting Multiple Resources

R. Dale et al. (Eds.): IJCNLP 2005, LNAI 3651, pp. 742 – 753, 2005. © Springer-Verlag Berlin Heidelberg 2005 Anaphora Resolution for Biomedical Literature by Exploiting Multiple Resources Tyne Liang and Yu-Hsiang Lin National Chiao Tung University, Department of Computer and Information Science, Hsinchu, Taiwan 300, ROC {tliang, gis91534}@cis.nctu.edu.tw Abstract. In this paper, a resolution system is presented to tackle nominal and pronominal anaphora in biomedical literature by using rich set of syntactic and semantic features. Unlike previous researches, the verification of semantic as- sociation between anaphors and their antecedents is facilitated by exploiting more outer resources, including UMLS, WordNet, GENIA Corpus 3.02p and PubMed. Moreover, the resolution is implemented with a genetic algorithm on its feature selection. Experimental results on different biomedical corpora showed that such approach could achieve promising results on resolving the two common types of anaphora. 1 Introduction Correct identification of antecedents for an anaphor is essential in message under- standing systems as well as knowledge acquisition systems. For example, efficient anaphora resolution is needed to enhance protein interaction extraction from biomedi- cal literature by mining more protein entity instances which are represented with pronouns or general concepts. In biomedical literature, pronominal and nominal anaphora are the two common types of anaphora. In past literature, different strategies to identify antecedents of an anaphor have been presented by using syntactic, semantic and pragmatic clues. For example, grammatical roles of noun phrases were used in [9] [10]. In addition to the syntactic information, statistical information like co-occurring patterns obtained from a corpus is employed during antecedent finding in [3]. However, a large corpus is needed for acquiring sufficient co-occurring patterns and for dealing with data sparseness. On the other hand, outer resources, like WordNet1, are applied in [4][12][15] and proved to be helpful to improve the system like the one described in [12] where ani- macy information is exploited by analyzing the hierarchical relation of nouns and verbs in the surrounding context learned from WordNet. Nevertheless, using Word- Net alone for acquiring semantic information is not sufficient for solving unknown words. To tackle this problem, a richer resource, the Web, was exploited in [16] 1 http://wordnet.princeton.edu/ Anaphora Resolution for Biomedical Literature by Exploiting Multiple Resources 743 where anaphoric information is mined from Google search results at the expense of less precision. The domain-specific ontologies like UMLS2 (Unified Medical Language System) has been employed in [2] in such a way that frequent semantic types associated to agent (subject) and patient (object) role of subject-action or action-object patterns can be extracted. The result showed such kind of patterns could gain increase in both precision (76% to 80%) and recall (67% to 71%). On the other hand, Kim and Park [11] built their BioAR to relate protein names to SWISS-Prot entries by using the centering theory presented by [7] and salience measures by [2]. In this paper, a resolution system is presented for tackling both nominal anaphora and pronominal anaphora in biomedical literature by using various kinds of syntactic and semantic …

Content truncated to 500 words