Cluster-based Language Model for Sentence Retrieval in Chinese Question Answering.pdf

Proceedings of the Fifth SIGHAN Workshop on Chinese Language Processing, pages 56–63, Sydney, July 2006. c?2006 Association for Computational Linguistics Cluster-based Language Model for Sentence Retrieval in Chinese Question Answering Youzheng Wu Jun Zhao Bo Xu National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences No.95 Zhongguancun East Road, 100080, Beijing, China (yzwu, jzhao,boxu)@nlpr.ia.ac.cn Abstract Sentence retrieval plays a very important role in question answering system. In this paper, we present a novel cluster-based language model for sentence retrieval in Chinese question answering which is mo- tivated in part by sentence clustering and language model. Sentence clustering is used to group sentences into clusters. Language model is used to properly rep- resent sentences, which is combined with sentences model, cluster/topic model and collection model. For sentence clustering, we propose two approaches that are One- Sentence-Multi-Topics and One- Sentence-One-Topic respectively. From the experimental results on 807 Chinese testing questions, we can conclude that the proposed cluster-based language model outperforms over the standard lan- guage model for sentence retrieval in Chinese question answering. 1 Introduction To facilitate the answer extraction of question answering, the task of retrieval module is to find the most relevant passages or sentences to the question. So, the retrieval module plays a very important role in question answering system, which influences both the performance and the speed of question answering. In this paper, we mainly focus on the research of improving the performance of sentence retrieval in Chinese question answering. Many retrieval approaches have been pro- posed for sentence retrieval in English question answering. For example, Ittycheriach [Ittycheriah, et al. 2002] and