Framework for multi-purpose web log access analyzer

In spite of being provided the abundant raw data from World Wide Web (WWW), it is difficult to make any sense out of such massive data without data mining techniques. In this paper, we emphasize on the mining of web access logs, web usage data. Our propose framework is composed of five steps. The first step is defining the purposes that our multipurpose analyzer can provide. In second step, we define the ontology mapping based on Web Sites defined in the previous step. After that, we perform preprocessing step base on our web log access data as the fundamental requirements of the proposed framework. Web usage based mining based on frequent item sets and our proposed algorithms are the core step of the system. In the final step, Naïve Bayesian classifier is applied to predict for the future depending on the current analysis outcomes. Our system is intended to provide for Web Site Maintainers, Web Site Developers, Personalization Systems, Pre-fetched Systems, Recommender Systems and Web Site Analysts, etc.