Bridging Lexical Gaps between Queries and Questions on Large Online Q&A Collections with Compact Translation Models

Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, pages 410–418, Honolulu, October 2008. c⃝2008 Association for Computational Linguistics Bridging Lexical Gaps between Queries and Questions on Large Online Q&A Collections with Compact Translation Models Jung-Tae Lee† and Sang-Bum Kim§ and Young-In Song‡ and Hae-Chang Rim† †Dept. of Computer & Radio Communications Engineering, Korea University, Seoul, Korea §Search Business Team, SK Telecom, Seoul, Korea ‡Dept. of Computer Science & Engineering, Korea University, Seoul, Korea {jtlee,sbkim,song,rim}@nlp.korea.ac.kr Abstract Lexical gaps between queries and questions (documents) have been a major issue in ques- tion retrieval on large online question and answer (Q&A) collections. Previous stud- ies address the issue by implicitly expanding queries with the help of translation models pre-constructed using statistical techniques. However, since it is possible for unimpor- tant words (e.g., non-topical words, common words) to be included in the translation mod- els, a lack of noise control on the models can cause degradation of retrieval performance. This paper investigates a number of empirical methods for eliminating unimportant words in order to construct compact translation mod- els for retrieval purposes. Experiments con- ducted on a real world Q&A collection show that substantial improvements in retrieval per- formance can be achieved by using compact translation models. 1 Introduction Community-driven question answering services, such as Yahoo! Answers1 and Live Search QnA2, have been rapidly gaining popularity among Web users interested in sharing information online. By inducing users to collaboratively submit questions and answer questions posed by other users, large amounts of information have been collected in the form of question and answer (Q&A) pairs in recent years. This user-generated information is a valu- able resource for many information seekers, because 1http://answers.yahoo.com/ 2http://qna.live.com/ users can acquire information straightforwardly by searching through answered questions that satisfy their information need. Retrieval models for such Q&A collections should manage to handle the lexical gaps or word mismatches between user questions (queries) and answered questions in the collection. Consider the two following examples of questions that are seman- tically similar to each other: • “Where can I get cheap airplane tickets?” • “Any travel website for low airfares?” Conventional word-based retrieval models would fail to capture the similarity between the two, be- cause they have no words in common. To bridge the query-question gap, prior work on Q&A retrieval by Jeon et al. (2005) implicitly expands queries with the use of pre-constructed translation models, which lets you generate query words not in a question by trans- lation to alternate words that are related. In prac- tice, these translation models are often constructed using statistical machine translation techniques that primarily rely on word co-occurrence statistics ob- tained from parallel strings (e.g., question-answer pairs). A critical issue of the translation-based ap- proaches is the quality of translation models con- structed in advance. If no noise control is conducted during the construction, it is possible for translation models to contain “unnecessary” translations (i.e., translating a word into an unimportant word, such as a non-topical or common word). In …

Content truncated to 500 words