计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (10): 132-136.

• 数据库、信号与信息处理 • 上一篇    下一篇

基于链接预测思想的问句检索方法研究

汪材印1,崔  琳2,李  鸿2,3   

  1. 1.宿州学院 机械与电子工程学院,安徽 宿州 234000
    2.宿州学院 信息工程学院,安徽 宿州 234000
    3.中国矿业大学 计算机科学与技术系,江苏 徐州 221008
  • 出版日期:2012-04-01 发布日期:2012-04-11

Question retrieval approach based on link prediction model

WANG Caiyin1, CUI Lin2, LI Hong2,3   

  1. 1.School of Mechanical Electric and Engineering, Suzhou University, Suzhou, Anhui 234000, China
    2.School of Information Engineering, Suzhou University, Suzhou, Anhui 234000, China
    3.School of Computer Science & Technology, China University of Mining & Technology, Xuzhou, Jiangsu 221008, China
  • Online:2012-04-01 Published:2012-04-11

摘要: 问答服务系统的一个重要功能是问题检索,即根据用户的提问,在已有的问答对数据中查找与用户提问相似的其他问题,将这些问题的答案直接返回给用户。问题检索任务所面临的主要困难是如何计算两个问句之间的语义相似度,提出利用链接预测模型计算问句之间的关联程度,将链接预测模型与语言模型相结合,设计出一种新的问句检索方法。通过在真实问答对数据上进行实验,表明该方法可以有效计算问句之间的语义相似度,其性能优于传统的计算方法。 

关键词: 问答系统, 链接预测, 信息检索

Abstract: In Q&A service, one of the important tasks is question retrieval. It means finding questions in the archive that are semantically similar to a user’s question that can satisfy the users’ need. The main challenge here is how to measure the semantic similarity between questions. In this paper, the link-prediction models are used to measure the latent relevance between questions. A model that combines the language model and the link-prediction model is proposed. The experiment results on a real Q&A data set show that it is possible to measure the semantic relationship between questions, and the approach outperforms the traditional calculating approaches.

Key words: Q&A system, link prediction, information retrieval