Navigational structure mining for usability analysis

Web hyperlink analysis has been successfully applied to improve search engines and to analyze patterns in the global Web. However, little work has been done in analyzing the hyperlink graph of a Web site for information architecture and usability analysis. In this paper, we propose a hyperlink frequent itemsets extraction algorithm, which allows the automatic extraction of navigational structures in a Web site without performing hypertext analysis. Experimental results show that the structure mining algorithm can automatically discover navigational structures within Web sites with high accuracy. Further tests with our novel usability system reveals that measured usability of a Web site have a direct correlation with properly designed navigational structures in the home of a Web site.