关键词: 用户访问模式, 电子商务, Web挖掘, 数据预处理

Abstract:

The comprehensive techniques of data collection and data preprocessing were discussed, which included data collection, data cleaning, user identification, session identification, episode identification and session sub-sequence generation. The methods for filtering frame, detecting search engine robots and generating user session semantic sequences, were proposed.

Key words: user access pattern, E-commerce, Web mining, data preprocessing