›› 2016, Vol. 29 ›› Issue (5): 130-.

• 论文 • 上一篇    下一篇

基于RTEM模型的问答社区候选答案排序方法

刘瑜,袁健   

  1. (上海理工大学 光电信息与计算机工程学院,上海 200093)
  • 出版日期:2016-05-15 发布日期:2016-05-24
  • 作者简介:刘瑜(1991—),女,硕士研究生。研究方向:数据挖掘、社区问答系统。袁健(1971—),女,博士,副教授。研究方向:数据挖掘、网络安全等。

Candidate Answer Sorting Method of Q&A Community Questions Based on RTEM Model

LIU Yu,YUAN Jian   

  1. (School of Optical-Electrical and Computer Engineering,University of Shanghai for
    Science and Technology,Shanghai 200093,China)
  • Online:2016-05-15 Published:2016-05-24

摘要:

问答社区作为搜索引擎的补充,在人们生活中起到越来越重要的作用,但随着大量问答对的积累,部分答案不能及时得到其他用户的评价,使得提问者选择最佳答案的难度加大。文中在研究用户行为及其行为所产生的结果后,提出一种新的基于主题模型的候选答案排序方法。通过大规模的实验数据证明,相较于其他模型方法,取得了较好的效果。

关键词: 问答社区, 答案排序, 主题模型, 用户行为

Abstract:

As a supplement of search engine,community question and answering is playing a more and more important role.But with increasing numbers of questions and answers,part of the answers fail to get the evaluation of other user in time,which makes it difficult for askers to choose the best answer.Based on the study of the user behavior of and its effects this paper proposes a new topic model on the basis of user behaviors.Experiments with large sets of data show that the proposed method is superior to other models.

Key words: community question and answering;answer ranking;topic model;user behavior

中图分类号: 

  • TP391