Different server parameters settings result in many different web log
types, but log files typically share the same basic information,
including client IP address, request time, requested URL, HTTP status
code, referrer, etc. Generally, several preprocessing tasks are required
before performing web usage mining algorithms on the
Web server logs. The tasks in this work include data cleaning, user
differentiation and session identification.