电子商务网站用户访问模式挖掘中的预处理技术

Preprocess techniques of mining user access patterns at E-commerce sites

对电子商务网站的用户访问模式挖掘中数据预处理阶段所采用的技术做了全面的研究,主要包括源数据的采集方法以及数据清理、用户识别、会话识别、事务识别、会话子序列生成等所采用的技术。并给出了框架网页过滤、识别搜索引擎Robot产生的访问记录,以及生成用户会话语义序列的方法。

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.