Novel Modified Fuzzy Possibilistic C Means (FPCM) for Web Log Mining by Removing Global Noise and Web Robots

Abstract

Nowadays, internet is a useful source of information in everyone’s daily activity. Hence, this made a huge development of world wide web in its quantity of interchange and its size and difficulty of websites. Web Usage Mining WUM is one of the main applications of data mining, artificial intelligence and so on to the web data and forecast the user’s visiting behaviors and obtains their interests by investigating the samples. SinceWUM directly involves in large range of applications, such as, e-commerce, e-learning, Web analytics, information retrieval etc. Web log data is one of the major sources which contain all the information regarding the users visited links, browsing patterns, time spent on a particular page or link and this information can be used in several applications like adaptive web sites, modified services, customer summary, pre-fetching, generate attractive web sites etc. There are varieties of problems related with the existing web usage mining approaches. Existing web usage mining algorithms suffer from difficulty of practical applicability. So, a novel research is very much necessary for the accurate prediction of future performance of web users with rapid execution time. The main aim of this paper to remove the noise and web robots by novel approach and provide faster and easier data processing and it also helps in saving time and it resource. In this paper, a novel pre-processing technique is proposed by removing local and global noise and web robots. Anonymous Microsoft Web Dataset and MSNBC.com Anonymous Web Dataset are used for evaluating the proposed preprocessing technique. An Effective Web User Analysis and Clustering are analyzed using Modified Fuzzy Possibilistic C Means (FPCM). Then results are evaluated using Hit Rate and Execution time.