图书情报工作 ›› 2016, Vol. 60 ›› Issue (18): 6-13.DOI: 10.13266/j.issn.0252-3116.2016.18.001

• 专题:基于OPAC日志挖掘的移动图书馆用户检索行为研究 • 上一篇    下一篇

移动图书馆用户查询式构造行为分析

吴丹, 董晶   

  1. 武汉大学信息管理学院 武汉 430072
  • 收稿日期:2016-08-10 修回日期:2016-09-03 出版日期:2016-09-20 发布日期:2016-09-20
  • 作者简介:吴丹(ORCID:0000-0002-2611-7317),教授,博士生导师,E-mail:woodan@whu.edu.cn;董晶(ORCID:0000-0002-6883-3901),硕士研究生。
  • 基金资助:

    本文系2016年度教育部人文社会科学研究规划基金项目“面向移动互联网的图书馆用户行为大数据分析与服务创新研究”(项目编号:16YJA870009)研究成果之一。

The Analysis of Mobile Library Users' Query Formulation Behavior

Wu Dan, Dong Jing   

  1. School of Information Management, Wuhan University, Wuhan 430072
  • Received:2016-08-10 Revised:2016-09-03 Online:2016-09-20 Published:2016-09-20

摘要:

[目的/意义] 揭示移动图书馆用户的查询式构造行为特征,并为移动图书馆的检索功能改进提出建议。[方法/过程] 采用系统日志挖掘法,根据某高校移动图书馆为期一个月的用户日志,通过统计分析方法,利用互信息值、查询式多样性、查询式丰富性、学科分布、持续时间等指标考察移动图书馆用户的查询式关联性、查询重构模式、查询式主题等方面。[结果/结论] 移动图书馆用户的查询式互信息值普遍较低,即查询式在内容上的关联性较弱;重复模式和直线模式是最常见的重构模式,即移动图书馆用户反复搜索同一查询式;移动图书馆用户的搜索兴趣集中在人文社科领域,用户对相同主题查询式的搜索行为具有持续性。建议增加查询推荐功能、自动纠错功能和高级检索功能,以提高移动图书馆检索服务的查全率和查准率。

关键词: 移动图书馆, 用户检索行为, 查询式构造, 日志挖掘

Abstract:

[Purpose/significance] The purpose of this research is to reveal characteristics of the query formulation behavior of mobile library users and to make some suggestions to improve the search function of the mobile library application. [Method/process] The characteristics in terms of the query relevance, query reformulation, query topics and request duration are examined by statistics of the mutual information, query diversity, query richness, topic distribution and duration time, based on the mobile library log mining of an academic library over a period of one month. [Result/conclusion] The mutual information statistics of mobile library users' queries are low, which means every query relates weakly semantically. The most common reformulation model are the Repetition in terms of adjacent queries and the Straight Line in terms of queries in a search session, which means mobile library users tend to search the same query for many times. It is found that mobile library users are interested in the information about humanities and social sciences, and they often search the same topic continuously. Thus, the paper suggests the search function of mobile library application add query recommendation, auto-correction, multiple search fields and advanced search.

Key words: mobile library, user seeking behavior, query formulation, log mining

中图分类号: