Web usage mining, data pre-processing, session identification, auxiliary pages, reference length, log files, business intelligence, data mining

Data mining belongs to the one of the important tools for Business Intelligence. It is a means toincrease competitiveness of a company. Web usage mining is engaged in data mining of webserver log fi le and it analyzes the user´s behavior on the web site. The fi rst step of web usagemining process is data pre-processing obtained from a web log fi le. Data pre-processing is animportant part of web usage mining. Discovering patterns of behavior of web visitors depends onthe quality of pre-processing phase. Therefore it is important to understand the used methods. Thispaper summarizes the pre-processing phases and especially the phases of session identifi cation.There are introduced two algorithms for data cleaning and session identifi cation using the referencelength method. The main aim of this paper is to compare a calculation of cutoff time and its infl uenceon discovered useful, trivial and inexplicable rules.