Navigation Analysis Tool based on the Correlation be- tween Contents Distribution and Access Patterns

This paper proposes a tool to assist web site publishers in improving their web site design. The tool uses an analysis technique that discovers the gap between web site publisher expectations and user behavior. The former are extracted by measuring the inter-page conceptual relevance and/or link connectivity. On the other hand, the latter is extracted by measuring the inter-page access co-occurrence. By evaluating the correlation between them, the tool discovers pages which should be improved in terms of web site design. It further uses a visualization technique using polar coordinate system to assist publishers in inspecting user access patterns in these pages. This technique will assist in finding clues for improving discovered gaps. The effectiveness of the tool is validated by case studies.