Web usage mining (WUM) is one of the types of data mining method which is used for analysing web usage patterns with the help of users' session and behavior. It is the technique to classify the web pages and internet users by taking into consideration the contents of the page and behavior of internet user in the past. Web mining extracts data accumulated in server access logs, referrer logs, agent logs, client-side cookies, user profile and Meta data. Usually the WUM techniques study the visitors' browsing behavior to obtain interesting knowledge. There are existing different techniques for web usage mining. Those existing techniques have their own advantages and disadvantages. Here we present a survey on some of the existing web usage mining techniques. First technique that we discuss is based on Sequential Pattern Algorithm. Sequential mining is the process of applying data mining techniques to a sequential database for the purposes of discovering the correlation relationships that exist among anordered list of events. Sequential mining techniques is web usage mining technique, where the sequences of web page accesses made bydifferent web users over a period of time, through a server, are recorded. Ford Lumban Gaol has proposed a web log sequential pattern mining using Apriori-all algorithm. Second technique that we discuss in this paper is based on distance between 2 sequences using no-Euclidean distance formula. Peiqian Liu and Wei Lihave proposed an improved Ward's methodfor web user clustering. They have given a formula, to calculate distance between elements which is a no-Euclidean distance measure. Since it is not a Euclidean distance the output preserves the ordering of events. In the last section of this document we have proposed a new web mining algorithm for analysing usage pattern of users. Proposed algorithm is based on WebLog Sequential Pattern Algorithm[Ford Lumban Gaol] and DBS[Peiqian Liu and Wei Li 2011]. )

Web Usage Mining Web Log sequential no Euclidean distance formula