Identification of Visitors Interest using Doubly Sparse Relevance Vector Machine (DSRVM) Classifier

In the area of research, web usage mining is certainly one of the upcoming techniques in domain of data mining. Since web contains millions of documents among them mining the data from the web is a difficult task. The problem defined is the extraction of patterns from web server log file leads to usage of mining using pattern recognition techniques. Therefore, the fundamental design of this approach is to determine a classifier that can achieve high accuracy for the most important class, visitors with and without purchase interest. By implementing Doubly Sparse Relevance Vector Machine (DSRVM) in web page classifier to improve precision and recall. Effective tracking enables the development and improvement of the user interface and software by analyzing user behavior. Based on fuzzy logic framework the proposed techniques is implemented for the identification of visitors. In this paper, the experimental results are compared with other classifiers such as SVM and RVM to prove the best outcome of DSRVM classifier.