[Purpose/significance]The user portrait label system that aggregates psychological preference labels can effectively improve the accuracy of user personalized services in university library.[Method/process]This paper obtains psychological preference satisfaction by integrating questionnaire survey,describes the new characteristics of university library user portrait based on user service scenarios,user attributes,user behaviors and other tags,and uses the random forest algorithm to train the model of user portrait recogni-tion.[Result/conclusion]The empirical analysis results show that the prediction accuracy of the model is 99.6％,and university li-brary can propose service marketing strategies such as optimizing service configuration,matching user characteristics and establishing feedback mechanism according to the important factors affecting service satisfaction,so as to effectively support university library to acti-vate data elements to enable business innovation and development.
关键词
用户画像/精准服务/随机森林/高校图书馆
Key words
user portrait/precise service/random forest/university library
维普
万方数据