图书情报工作 ›› 2021, Vol. 65 ›› Issue (23): 126-135.DOI: 10.13266/j.issn.0252-3116.2021.23.014

• 知识组织 • 上一篇    下一篇

融合Word2vec和WGRA的社会化问答社区答案有用性排序方法研究——以携程问答为例

郭顺利, 步辉   

  1. 曲阜师范大学传媒学院 日照 276800
  • 收稿日期:2021-06-23 修回日期:2021-09-22 出版日期:2021-12-05 发布日期:2021-12-18
  • 作者简介:郭顺利,讲师,博士,E-mail:guosl777@163.com;步辉,硕士研究生。
  • 基金资助:
    本文系国家社会科学青年基金项目"基于认知计算的网络问答社区知识的深度聚合及精准服务研究"(项目编号:20CTQ028)研究成果之一。

Research on the Sorting Method of Answer Usefulness in Social Q&A Community Integrating Word2vec and WGRA-Taking Ctrip Q&A as an Example

Guo Shunli, Bu Hui   

  1. School of Communication, Qufu Normal University, Rizhao 276800
  • Received:2021-06-23 Revised:2021-09-22 Online:2021-12-05 Published:2021-12-18

摘要: [目的/意义]为解决社会化问答社区用户信息需求多样化和答案冗余过载问题,提出面向用户个性化需求的答案有用性排序方法,协助用户高效筛选和获取有用的答案知识。[方法/过程]首先通过文献调研和专家咨询法,从答案特征、回答者特征、答案的时效性3个维度构建答案有用性评价指标体系;然后,从语义层面融合用户个性化需求,设计融合加权灰色关联分析法和Word2vec的答案有用性排序方法,实现面向用户需求的答案排序。[结果/结论]通过实验结果的对比分析发现与基于"点赞数"和"回答时间"等传统的排序方法相比,笔者设计的答案有用性排序方法的用户满意度更高,更能够满足用户的个性化知识需求。

关键词: 用户需求, 答案有用性, 加权灰色关联度, Word2vec, 社会化问答社区

Abstract: [Purpose/significance] In order to solve the diversified information needs of users and the problem of redundant and overloaded answers in the social Q&A community, this paper proposes an answer usefulness ranking method oriented to users' personalized needs,assists users to efficiently filter and obtain useful answer knowledge.[Method/process] First, through literature research and expert consultation, an answer usefulness evaluation index system was constructed from the three dimensions of answer characteristics, answerer characteristics and answer timeliness; Then, it integrated the user's personalized needs from the semantic level, designed an answer usefulness ranking method that combined WGRA and Word2vec, and realized the answer ranking oriented to user needs.[Result/conclusion] Through comparative analysis of experimental results, it is found that compared with traditional ranking methods based on "likes" and "answer time", the answer usefulness ranking method designed in this paper has higher user satisfaction and is more able to satisfy users' personalized knowledge demands.

Key words: user demand, answer usefulness, WGRA, Word2vec, social Q&A

中图分类号: