Paper Title
A Comprehensive Study of Genetic Algorithms for Classification of Browsing Behavior Data

Abstract
This paper makes a study about genetic algorithms of classification rule mining on browsing behavior data of surfers. Genetic algorithms are evolutionary approaches and are suitable for large search spaces as they cope better with attribute interaction than greedy rule induction algorithms. Web usage database contains huge data about patterns of usage of web elements. These patterns can be classified into various categories of tasks. This paper presents a comparative study of various Genetic algorithms of classification with that of standard efficient greedy algorithms on benchmark data as well as real data of browsing behavior of surfers. In order to increase the efficiency of the process, the data have been pre-processed with efficient algorithms of discretization and feature selection. Results are encouraging and give important insight about the browsing behavior data. It shows that genetic algorithms are more accurate on benchmark data but lack this performance on the used web usage data. Keywords - Genetic Algorithms of Classification Rule Mining, Browsing Behavior, Web Usage Mining.