Research on mining user browsing patterns in large web logs based on Poisson sampling and Sequence Alignment Method

The continuous growth in the size and use of the Internet is creating large sever logs on web servers and difficulties in the search for information. A sophisticated method to organize the layout of the information and assist user navigation is therefore particularly important. In this paper, we valuate the feasibility of using a Poisson sampling and SAM to mine large web log data. A sample sets selected by Poisson sampling statistically effectively represent the characteristics of the entire dataset. In addition, users are partitioned into clusters using a non-Euclidean distance measure, called Sequence Alignment Method (SAM).