Open AccessAn Empirical Study on the Phenomenon of Information Narrowing in the Context of Personalized Recommendation
Author(s) -
Yijun Huang,
Lan Zhou,
Ziqian Zeng,
Duan Ling-li,
Jiayu Wang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1631/1/012109
Subject(s) - computer science , context (archaeology) , empirical research , information overload , phenomenon , perception , information retrieval , object (grammar) , information filtering system , world wide web , data science , psychology , artificial intelligence , statistics , paleontology , physics , mathematics , quantum mechanics , neuroscience , biology
Personalized recommendation services have been widely applied to improve users’ search efficiency, but when filtering information for users, they also hide the risk of the declining in content diversity, which is information narrowing. In this study, comprehensive news and information platforms in China were taken as the empirical research object. The experimental tracking method was used to collecting the content data of client end recommended news. Simpson and Shannon-Wiener indexes were used to measure the information narrowing phenomenon. Questionnaires were used to collect users’ data of news and information applications usage, the perception of information overload, and the perception of information narrowing. The data of questionnaire was analyzed and processed by MATLAB and SPSS 20.0, and the hypothesizes were tested by statistical methods. The results reveal that there is indeed information narrowing phenomenon in news information personalized recommendation service, but most users are unaware of this.