Analyzing User Behavior History for Constructing User Profile

摘要：

User Behavior History has been employed to discover users preferences and interests for personalized systems. In this paper we propose a framework for extracting user interested items by analyzing user behavior history in digital library environment. Firstly, this paper models every users behaviors as a User Behavior Forest (UBF). Then user behavior log is formatted with several additional attributes, and we show how to construct UBF from such logs. At last, an algorithm is presented for measuring user interests on items by considering users behavior sequence, users behavior type, and the metadata of literature.

作者: CHEN Yu YU Yang ZHANG Wei SHEN Junyi

作者单位: Beijing Document Service,Beijing,100142,China;School of Electric and Information Engineering,Xian J Beijing Document Service,Beijing,100142,China School of Electric and Information Engineering,Xian Jiaotong Univ.,Xian,710049,China

会议类型: 国际会议

会议名称: 2008 IEEE International Symposium on IT in Medicine and Education(2008信息技术在医学和教育中的应用国际研讨会)(ITME 2008)

会议地点: 厦门

会议语种:英文

页码: 343-348

在线出版日期: 2008-12-12（万方平台首次上网日期，不代表论文的发表时间）