Text Data Mining of User Reviews in"Double First-class"University Libraries Based on LDA Topic Model
[Purpose/significance]The library user reviews contain important information about users'opinions on library services and management.The in-depth analysis of user review texts aims to explore the topics users are concerned about,their emotional atti-tudes,and implicit demands,so as to provide theoretical basis and data support for the construction of university libraries.[Method/process]This paper takes the"double first-class"university libraries as the research objects,collects user reviews data on Dianping.com,and through LDA topic modeling,gets that the current user reviews mainly focus on three dimensions:information resource con-struction,entertainment and leisure services,and library facilities and environment.The keyword co-occurrence network and users'e-motional attitude under the three dimensions are then analyzed separately.[Result/conclusion]The overall emotional attitude of users to university libraries presents positive,but they show negative emotions in paper resource management,social services,and humanized services in the library.
university libraryuser reviewsLDA topic analysisco-occurrence networksentiment analysis
NETL
NSTL
万方数据
