This paper introduces a new tool for web usage mining and visualization that relies on the bio-mimetic relational clustering algorithm Leader Ant and the definition of pro- totypes based on typicality computation to produce an effi- cient visualization of the activity of users on a website. The tool is evaluated on a real web log file from the French mu- seum of Bourges website and shows that it can easily pro- duce meaningful visualizations of typical user navigation.
Citation:
Nicolas Labroche, Marie-Jeanne Lesot, Lionel Yaffi, "A New Web Usage Mining and Visualization Tool," ictai, vol. 1, pp.321-328, 19th IEEE International Conference on Tools with Artificial Intelligence - Vol.1 (ICTAI 2007), 2007