An incremental SOM for web navigation patterns clustering
Khalid Benabdeslem, Younès Bennani · HAL (Le Centre pour la Communication Scientifique Directe) · 2006
In this paper, we present a new clustering method which makes incremental the construction of an unsupervised neural model (Self Organizing Map: SOM). In other words, the method is computed with both, the initial model based on the a priori available data and the data which arrive dynamically in the time. This approach is validated over web navigation data and it is compared to classical neural clustering applied to the same data.