A new algorithm based on cellular learning automata for web usage mining

One of the methods of data mining on the web, is web usage mining. The purpose of the mining, discovery of useful information from the data obtained from the contact of users behavior while the users interact with the web. When users visit web pages for specific sequences, this represented a kind of relationship between them. These sequences are called navigation patterns of users, and the mining performed on the navigation patterns is called mining navigation patterns of users. These navigation patterns during the process of sessionization with mining of log files are obtained. In this paper, we propose a new algorithm based on cellular learning automata to determine the relationship between web pages. Determining the relationship between the structures of web pages makes find pages that are similar to each other, by which we can do clustering and ranking web pages. For this purpose, the proposed algorithm uses user navigation patterns, which are saved on log files. In order to study the efficiency of the proposed algorithm, the computer simulations are conducted. The simulation results show that the efficiency of the proposed algorithm compared to the existing algorithms is high. This efficiency by comparison with existing algorithms will be evaluated.

A new algorithm based on cellular learning automata for web usage mining Keywords:

web usage mining , user navigation pattern , structure of web pages , cellular learning automata