Multilingual Robust Anaphora Resolution

Multilingual robust anaphora resolution Ruslan Mitkov School of Languages and European Studies University of Wolverhampton Stafford Street Wolverhampton WVJ ISB, United Kingdom Email R.Mitkov@wlv.ac.uk Lamia Belguith LARIS -FSEG University of Sfax B.P. 1088 3018 Sfax, Tunisia Email belguith.lamia@planet.tn Malgorzata Stys Computer Laboratory University of Cambridge New Museums Site, Pembroke Street Cambridge CB2 3QG United Kingdom Email Malgorzata.Stys@cl.cam.ac.uk Abstract Most traditional approaches to anaphora resolution rely heavily on linguistic and domain knowledge. One of the disadvantages of developing a knowledge-based system, however, is that it is a very labour-intensive and timehconsuming task. This paper presents a ro- bust, knowledge-poor approach to resolving pronouns in technical manuals. This approach is a modification of the practical approach (Mitkov 1998a) and operates on texts pre-processed by a part-of-speech taggcr. In- put is checked against agreement and a nlonhber of an- tecedent indicators. Candidates are assigried scores by each indicator and the candidate with the highest ag~ gregate score is returned as the antecedent. We pro~ pose this approach as a platform for multilingual pro- noun resolution. The robust approach was initially de~ veloped and tested for English, but we have also adapted and tested it for Polish and Arabic. For both languages, we found that adaptation required mini- mum modification and that further, even if used un~ modified, the approach delivers acceptable success rates. Preliminary evaluation reports high success rates in the range of and over 90% 1. Introduction: robust, knowledge poor anaphora resolution and multilingual NLP For the most part, anaphora resolution has focused on traditional linguistic methods (Carbonell & Brown 1988; Carter 1987; Hobbs 1978; Ingria & Stallard 1989; Lappin & McCord 1990; Lappin & Leass 1994; Mitkov 1994; Rich & LuperFoy 1988; Sidner 1979; Webber 1979). However, to represent and manipulate the various types of linguistic and domain 7 knowledge involved requires considerable human input and computational expense. While various alternatives have been proposed, making use of e.g. neural networks, a situation se- mantics framework, or the principles of reasoning with uncertainty (e.g. Connoly et a!. 1994; Mitkov 1995; Tin & Akman 1995), there is still a strong need for the development of robust and effective strategies to meet the demands of practical NLP systems, and to enhance further the automatic processing of growing language resources. , Several proposals have already addressed the anaphora resolution problem by deliberately limiting the extent to which they rely on domain and/or lin- guistic knowledge (Baldwin 1997; Dagan & llai 1990; Kennedy & Boguraev 1996; Mitkov 1998; Nasukawa 1994; Williams eta!. 1996). Our work is a continuation of these latest trends in the search for inexpensive, rapid and reliable procedures for anaph~ ora resolution. It shows how pronouns in a specific genre can be resolved quite successfully without any sophisticated linguistic knowledge or even without parsing, benefiting instead from corpus-based NLP techniques such as sentence splitting and part-of- speech tagging. On the other hand, none of the projects reported so far, has looked at the multilingual aspects of the approaches that have been developed, or, …

Content truncated to 500 words