05838373

a

05838373 Yang, Peng Zheng, Yan Inspired rule-based user identification. Cao, Longbing (ed.) et al., Advanced data mining and applications. 6th international conference, ADMA 2010, Chongqing, China, November 19--21, 2010. Proceedings, Part I. Berlin: Springer (ISBN 978-3-642-17315-8/pbk). Lecture Notes in Computer Science 6440. Lecture Notes in Artificial Intelligence, 618-624 (2010). 2010 Berlin: Springer EN Web Log Mining Preprocessing User Identification

doi:10.1007/978-3-642-17316-5_60

Summary: Web log mining is an important branch of Web usage mining. It is a kind of specific application of data mining. It can find Web user mode of behavior through processing server-side log files, and further improve the structure of the website or provide users with personalized service. Data preprocessing is an important step of Web log mining. In general, data preprocessing is divided into four steps: data cleaning, user identification, session identification, path supplement, transaction identification. This paper proposes a user identification method which is based on inspired rules. Experiment results prove the effectiveness of this method.