Feature Selection Optimization using ACO to Improve the Classification Performance of Web Log Data

Web mining plays a salient role in today’s era of an interconnected network to comprehend the manner in which users browse the websites. Web mining can be classified into three main categories such as web content mining, web structure mining, and web usage mining. Web structure mining deals with the analysis of the structure of web sites, web pages while web content mining deals with the analysis of the contents on the web pages. Web usage mining processes web server log files for the applications of various functionalities such as classification, clustering, and association rules on weblog data. In this paper, the authors presented the classification of data on the server log files by applying various popular and most widely used approaches of classification on the dataset. Then, based on the highest predictive accuracy, the random forest algorithm was further optimized using ant colony optimization (ACO) techniques and the authors have observed that the classification accuracy can be improved using ACO on the results of the random forest algorithm.