期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2016
卷号：5
期号：11
页码：2591-2595
出版社：Shri Pannalal Research Institute of Technolgy
摘要：This paper, introduces “Segmentation of User’sTask” to understand user search behavior. Today the massiveuse of the web has drastically increased the traffic in thenetwork together with the load that web servers need tomanage. Sometime user leaves the sites because of its highlatency time, so this becomes one important subject forresearch. Web mining is considered to store the information ina specific format known as weblog. This important feature haslot of implementation such as query recommendation, querylog analysis, and many other techniques for improvement. Asthe query classification is the main motive of this paper whereone can generate the class of the query for providing betterresponse of the query. This paper proposes a model whereproper knowledge from the ontology and the web usage of theweb feature vector are created for training the Error backpropagation neural network. This model will increase theaccuracy of the classification, so that web server response willbe small. Here overall execution time will be reduced becausethe trained neural network runs in constant time.
关键词：Information Extraction; web log; web;query ranking; web mining.
