基于用户生成内容的绘本资源分类策略研究——以小红书与豆瓣平台为例  

Research on the Resource Classification Strategies of Picture Books Based on User Generated Content——Taking Xiao hongshu and Douban Platform for Example

作　　者：李芙蓉 高璇 Li Furong;Gao Xuan

机构地区：[1]上海大学文化遗产与信息管理学院,上海200444

出　　处：《山东图书馆学刊》2024年第6期39-44,共6页The Library Journal of Shandong

摘　　要：通过分析UGC平台上由用户生成的绘本标签与短评,以期构建一个综合分类系统用于绘本资源分类。通过手动筛选,获取到小红书平台绘本标签1375个,使用八爪鱼采集器采集豆瓣平台绘本短评73472条,进一步使用Python与微词云文本分析工具对绘本标签、短评数据进行文本内容挖掘分析,对绘本标签进行标签长度、词性与共现网络分析,对绘本短评进行高频特征词、LDA主题聚类与情感倾向性分析。从中提炼出绘本标签的五个内容特征,进一步凝练出绘本短评的五个主题,最终提出了优化绘本资源分类的三个策略构想。Through the analysis of user-generated picture book tags and short comments on UGC platform,it is expected that the popular classification method and the traditional classification method can be combined to classify picture books in picture book resource service institutions such as libraries.Through manual screening,1375 picture book tags on Xiaohongshu were obtained,73,472 short comments on Douban platform were collected by Bazhuayu,and text mining and analysis of picture book tags and short book comments data were further carried out by using Python and Weiciyun text analysis tools,as well as tag length,tag part of speech and tag co-appearance network analysis of picture book tags.The paper analyzes the high frequency feature words,LDA theme clustering and emo-tional tendency of picture book short comments.The paper extracts five content characteristics of picture book tags,further summarizes the five themes of picture book short comments,and finally puts forward three ideas of picture book resource classification.

关 键 词：社会化标签 短评 文本挖掘 绘本资源分类 

分 类 号：G254.1[文化科学—图书馆学]