摘要
[目的/意义]更全面地理解用户标签使用行为的影响因素,为标签推荐和应用实践提供指导。[方法/过程]采用日志分析法,从人均标签个数、标签类型分布两个方面分析图书主题的影响,并以计算机、心理学、经济学、文学作品、绘画5类图书为例,基于社会化标注社区豆瓣网的用户数据进行实证分析。[结果/结论]图书主题会对用户的标签个数及类型分布具有显著影响:1在标签个数方面,计算机、心理学、经济学、文学作品主题图书的人均标签个数相近,显著高于绘画主题图书;2在标签类型分布上,根据5类图书间的差异可分为3组:计算机、心理学和经济学、文学作品和绘画。这一结论对图书标签推荐及图书分面检索和导航系统的设计具有参考作用。
[ Purpose/significance ] This paper aims to fully understand the influence factors of users＇ social tagging behaviors and provide guidance for the tag recommendation. [ Method/process] Using the method of log file analysis, this paper analyzes the influence of book subjects from the perspectives of average number of tags per user and category distri- bution of tags. Then it makes an empirical analysis on users data of Douban community by taking five kinds of books as samples, including computer, psychology, economics, literature and painting. [ Result/conclusion] The results show that the book subject has significant influence on users＇ number and category distribution of tags ：（1）for the number of tags, the average number per user on books in subjects of computer, psychology, economics and literature are similar, while higher than that of books in subject of painting;（2）for the category distribution of tags, based on the difference of tags category distribution, the five kinds of books are divided into three groups of computer, psychology and economics, literature and painting. These conclusions have important guidance for the book tags recommendation, and the books faceted search and navigation system design.
出处
《图书情报工作》
CSSCI
北大核心
2016年第8期106-112,共7页
Library and Information Service
基金
教育部人文社会科学基金项目"基于企业创新价值链的知识融合服务研究"(项目编号:15YJA870005)研究成果之一
