Performance Comparison of Tree-Based Machine Learning Classifiers for Web Usage Mining

Web usage mining plays a very important role in finding a new patterns and recognition from the web server log file. It is the subcategory of web mining that is used to mine data over the web. In this paper, the authors classified the server log file, identified new patterns of data, and analyzed the data over the tree-based classification algorithms. Authors accessed the tree algorithms (treeJ48, RandomTree, RandomForest, and REPTree) to classify the weblog data and analyzed the results. This paper summarizes the results and compared the tree-based algorithms classification data and check which algorithm gives a better result.