Enhancement searching on internet using web data mining techniques

The rapid increase in web sites has created lar~e volume of data inweb environment, which generate a problem of how to extract and gainuseful knowledge from such data. Extracting such knowledge introducesa rapid services, saves user time with web, and improves user’s efficiencyand effectiveness in searching for information on the web. Also thisknowledge helps to discovering, and predicting user’s behaviors based ontheir interaction with a website (Personalization).In this thesis an overview of approaches for data mining and itsobjectives, process and applications would be explained. Web miningconcept (web content mining, web structure mining, web usage mining)and its applications would be introduced .General framework for fullystages of web usage mining including (preprocessing, pattern discoveryand pattern analysis), and an approach for e-business and personalizationoperation would be discussed. Also an overview for different types ofdata sources, and a brief description for web log files structures would bediscussed.Web muung techniques used for analyzing web data likeClassification ,Clustering, Sequential Patterns, Association Rules ,andspecially Artificial Neural Network behaviors (Operation, Architecture,Different Training Rules Algorithms and Applications) would bediscussed.The aim of this thesis is analyzing web log file data for InformationSystem of Mansoura University Center using proposed Web Knowledge”Extraction from Logs Files (WKELF) analyzing system .This analysis