Improvised hybrid model for user navigation pattern prediction

Abstract

Web usage mining involves identifying the user's navigation patterns from the web server log files. These patterns can be further analyzed by applying certain data mining algorithms to it. The discovered navigation patterns can be further used for several things like identifying the frequent patterns of the user, predicting the future request of user, etc. It can be used in the fields of recommendation, web caching, improvement of web site personalization, etc. A hybrid approach is being proposed in this paper which involves a combination of K-harmonic means algorithm (KHM) for grouping the identified patterns and Frequent Pattern (FP) growth algorithm for identifying frequent patterns. The results of the proposed work are compared against the existing systems.