Data Preprocessing: A Milestone of Web Usage Mining

-.Internet is today full of structured or unstructured information. and this information is directly or indirectly influencing society or peoples. Because today internet is part our daily life activity. But using this abundant and ambiguous in most efficient manner in useful decision making is still a big challenge. During our web surfing either it is online shopping or blogging or using tweets and chatting everything is recorded. Web servers and log files are used to collect all the activity information of a user accesses to a web server. Log data is usually noisy and ambiguous. Web usage mining, also known as web log mining, is the application of data mining techniques on large web log databases to discover knowledge about user behavior pattern and web site usage statistics that can be used for various website design tasks. WUM consists of three phases:-Preprocessing, Pattern discovery and Pattern analysis. It is a fact that normal log files data is very huge, noisy, unclear and confusing with lots of redundancy. It is very important to preprocess the log data for efficient web usage mining process. Preprocessing results also influences the later phases of web usage mining. This makes the preprocessing of server log files a significant step in web usage mining. This study includes analysis, comparison and contrast of the available preprocessing techniques. How we can be more focused and guided at preprocessing level. So in this paper, we have given a complete preprocessing analysis by reviewing the existing work done in the preprocessing stage.