International Journal of Engineering and Technology

Clustering of web user sessions is extremely significant to comprehend their surfing activities on the internet. Users with similar browsing behaviour are grouped together, and further analysis of discovered user groups by domain experts may generate usable and actionable knowledge. In this paper, a conglomerative clustering approach is presented to identify web user session clusters from web server access logs, based on their browsing behaviour. Presented algorithm captures essential ideas from subtractive and relational fuzzy c-mean clustering algorithm.This algorithm works in two phases, in the first phase, it automatically identifies the number of potential clusters based on the successively subtractive potential density function value of each relational data and their respective centres (centroid). In the second phase, it assigns fuzzy membership values to from fuzzy clusters from a relational matrix. The presented algorithm is applied on an augmented session dissimilarity matrix obtained from an openly accessible NASA web server log data..