Mining Web Navigation Patterns with Dynamic Thresholds for Navigation Prediction

Discovering web navigation patterns is an important issue in web usage mining with various applications like navigation prediction and improvement of website management. Since web site structure is always changed, we need not only consider the frequency of click behavior but also web site structure to mine web navigation patterns for navigation prediction. To reduce the overhead of dynamically mining the web navigation patterns from the web data, a dynamic mining approach is needed by using the previous mining results and computing new patterns just from the inserted or deleted part of the web data. In this paper, we propose a special data structure named Ideal-Tree (Inverted-database Expectable Tree) to avoid the effort of scanning database. Meanwhile, an efficient mining algorithm named Ideal-Tree-Miner is proposed for mining web navigation patterns with dynamic thresholds. Based on the discovered patterns, we also give a navigation prediction model. The experimental results show that our prediction model outperforms other approaches substantially in terms of Precision, Recall, and F-measure.