中国邮电高校学报(英文) ›› 2014, Vol. 21 ›› Issue (1): 116-121.doi: 10.1016/S1005-8885(14)60277-6

• Others • 上一篇    下一篇

Ensemble similarity measure for community-based question answer

孙月萍1,王小捷1,王序文1,姜邵巍1,LIU Yong-bin   

  1. School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • 收稿日期:2013-07-10 修回日期:2013-12-30 出版日期:2014-02-28 发布日期:2014-02-28
  • 通讯作者: 孙月萍 E-mail:moonpingsun@126.com
  • 基金资助:

    This work was supported by the National Natural Science Foundation of China (61273365), the National High Technology Research and Development Program of China (2012AA011104).

Ensemble similarity measure for community-based question answer

  1. School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2013-07-10 Revised:2013-12-30 Online:2014-02-28 Published:2014-02-28
  • Contact: Yue-Ping SUN E-mail:moonpingsun@126.com
  • Supported by:

    This work was supported by the National Natural Science Foundation of China (61273365), the National High Technology Research and Development Program of China (2012AA011104).

摘要:

Community-based question answer (CQA) makes a figure network in development of social network. Similar question retrieval is one of the most important tasks in CQA. Most of the previous works on similar question retrieval were given with the underlying assumption that answers are similar if their questions are similar, but no work was done by modeling similarity measure with the constraint of the assumption. A new method of modeling similarity measure is proposed by constraining the measure with the assumption, and employing ensemble learning to get a comprehensive measure which integrates different context features for similarity measuring, including lexical, syntactic, semantic and latent semantic. Experiments indicate that the integrated model could get a relatively high performance consistence between question set and answer set. Models with better consistency tend to get a better precision according to answers.

关键词:

similar question retrieval, similarity measure, CQA, ensemble learning

Abstract:

Community-based question answer (CQA) makes a figure network in development of social network. Similar question retrieval is one of the most important tasks in CQA. Most of the previous works on similar question retrieval were given with the underlying assumption that answers are similar if their questions are similar, but no work was done by modeling similarity measure with the constraint of the assumption. A new method of modeling similarity measure is proposed by constraining the measure with the assumption, and employing ensemble learning to get a comprehensive measure which integrates different context features for similarity measuring, including lexical, syntactic, semantic and latent semantic. Experiments indicate that the integrated model could get a relatively high performance consistence between question set and answer set. Models with better consistency tend to get a better precision according to answers.

Key words:

similar question retrieval, similarity measure, CQA, ensemble learning