利用双语翻译对社区问答进行问题检索.PDFVIP

Exploiting Bilingual Translation for Question Retrieval in Community-Based Question Answering Guangyou Zhou, Kang Liu and Jun Zhao National Laboratory of Pattern Recognition Institute of Automation, Chinese Academy of Sciences 95 Zhongguancun East Road, Beijing 100190, China {gyzhou, kliu, jzhao@ ABSTRACT Community-based question answering (CQA) has become an important issue due to the pop- ularity of CQA archives on the web. This paper is concerned with the problem of question retrieval. Question retrieval in CQA archives aims to find historical questions that are seman- tically equivalent or relevant to the queried questions. However, question retrieval is chal- lenging partly due to the word ambiguity and lexical gap between the queried questions and the historical questions in the archives. To deal with these problems, we propose the use of translated words to enrich the question representation, going beyond the words in the original language to represent a question. In this paper, each original language question (e.g., English) is automatically translated into an foreign language (e.g., Chinese) by machine translation ser- vices, and the resulting translated questions serves as a semantically enhanced representation for supplementing the original bag of words. Experiments conducted on real CQA data set demonstrate that our proposed approach significantly outperforms several baseline methods and achieves the state-of-the-art performance. TITLE AND ABSTRACT IN ANOTHER LANGUAGE, L2 (OPTIONAL, AND ON SAME PAGE) 利利利用用用双双双语语语翻翻翻译译译对对对社社社区区区问问问答答答进进进行行行问问问题题题检检检索索索 由于互联网上社区问答数据集的流行，使得社区问答的研究变得越来越流行。本文 关注的是问题检索。问题检索的目的是从历史问题数据集中查找与查询问题语义等价或相 关的历史问题。然而，问题检索的挑战主要是词汇歧义和查询问题与历史问题之间的词汇 鸿沟。为了解决这些问题，我们提出利用翻译词来丰富问题的表示，而不单纯利用原始语 言的词来表示问题。在本文中，通过机器翻译，每个原始语言（例如