High Utility Web Access Patterns Mining from Distributed Databases

Utility based web access pattern mining and knowledge discovery from database has become an interesting research domain in recent time. Traditional pattern mining algorithms deal with only binary occurrence of a web page and do not consider the weight or profit of the web page. Hence, utility-based web path traversal pattern mining technique has evolved and got much interest in recent time. However, the methods of mining high utility web access pattern from distributed databases have not explored yet. All of the existing utility pattern mining algorithms are based on centralized database. In this paper, we have been proposed a parallel and distributed method for mining high utility web access patterns. We also proved that our algorithm maintains downward closure property. Extensive experimental analyses reflect that our proposed technique is highly efficient on large databases in distributed environment