Survey on recent methodologies used for recommender system
Prathamesh S Tugaonkar, Vidya Chitre · 2016
Internet is an enormous store of links and web pages that provides huge information to users. The development rate of web is as about one million pages a day. Users' points are traced in web logs. Hence, its obvious that number of log files created is also large. This paper give information of Web usage mining which generally has three phases as first is preprocessing, next pattern detection and last pattern learning. Web log data is noisy and confusing, thus last two phases are essential. Traditional data processing system has bottleneck in knowledge storage and computing power. Use of Cloud may help us to reduce such issues in mining._This paper will give information regarding different stages in Web Usage Mining and Problems with related recent advancement in this research area. Cloud Mining with Hadoop and Map Reduce is also introduced.