An Improved Web Log Data Preprocessing
Liang Tang · Techniques of Automation and Applications · 2008
There is massive redundant information in Web log,which can lead to inconvenience for data mining.Data preprocess-ing is to treat these redundant information such as data purifying,user recognition,session recognition and pathway complement.This data preprocessing is introduced in detail.Meanwhile,the traditional session recognition algorithm is also improved.The system is based on the SQL sentence.