图书情报工作 ›› 2019, Vol. 63 ›› Issue (21): 66-77.DOI: 10.13266/j.issn.0252-3116.2019.21.007

• 情报研究 • 上一篇    下一篇

基于LDA模型因素提取的健康信息用户转移行为研究

张泰瑞, 陈渝   

  1. 昆明理工大学管理与经济学院 昆明 650093
  • 收稿日期:2019-01-09 修回日期:2019-04-11 出版日期:2019-11-05 发布日期:2019-11-05
  • 通讯作者: 陈渝(ORCID:0000-0003-4954-0406),教授,博士生导师,通讯作者,E-mail:cykust@126.com
  • 作者简介:张泰瑞(ORCID:0000-0002-5591-6708),硕士研究生。
  • 基金资助:
    本文系国家自然科学基金项目"基于示能性视角的信息系统有效使用研究:维度、影响因素和形成机制"(项目编号:71461016)和国家自然科学基金项目"医联体内电子健康档案垂直整合研究"(项目编号:71864021)研究成果之一。

A Research on User Transfer Behavior of Health Information Based on LDA Model Factor Extraction

Zhang Tairui, Chen Yu   

  1. School of Management and Economics, Kunming University of Science and Technology, Kunming 650093
  • Received:2019-01-09 Revised:2019-04-11 Online:2019-11-05 Published:2019-11-05

摘要: [目的/意义]移动健康信息服务的发展可以为医疗行业创造巨大的潜在价值,但整体应用滞后于金融服务等行业,其源于转移过程中存在诸多的影响因素。对用户健康信息行为研究成果进行文本挖掘能够有效揭示潜在的主要影响因素,进而总结这些因素对转移行为的影响。[方法/过程]利用LDA模型进行文本分析,结合社会认同理论,将相对感知价值、相对忠诚度作为关键影响因素,转移成本作为调节因素,构建用户转移行为影响因素的研究模型并进行实证分析。[结果/结论]研究结果显示,LDA模型分析结果可以获取较为具体的特征词,通过分析特征词能够获取有效的影响因素,同时,这些因素与实际理论具有较高契合度,基于社会认同理论与相关因素理论提出的模型对用户转移行为具有较好的解释效果。

关键词: 健康信息服务, LDA模型, 因素探索, 用户转移行为

Abstract: [Purpose/significance] The development of mobile health information services can create huge potential value for the medical industry, but the attitude of users is vague, which stems from many influencing factors in the process of transfer. Text mining of user health information behavior research results can effectively reveal the main potential influencing factors, and then it summarizes the impact of these factors on the transfer behavior.[Method/process] Using LDA model for text analysis, combined with social identity theory, the relative perceived value and relative loyalty as the key influencing factors, the transfer cost as the moderating factor, the research model of the influencing factors of user transfer behavior is constructed and empirically analyzed.[Result/conclusion] The results show that the LDA model analysis results can obtain more specific feature words, and the effective influence factors can be extracted by analyzing the feature words. At the same time, these factors have a high degree of fit with the actual theory. The proposed structural equation model has a good interpretation effect on user transfer behavior.

Key words: health information service, LDA model, factor exploration, user transfer

中图分类号: