Multidimensional Sessions Comparison Method Using Dynamic Programming

This paper introduces a new multidimensional sessions comparison method (MSCM) using dynamic programming. Our method takes into consideration of different session dimensions such as the page list, the time spent on each page, and the length of each session. The method showed more accurate results than other known methods such as sequence alignment method (SAM), multidimensional sequence alignment method (MDSAM), and path feature space. Output of the MSCM is presented in the form of dissimilarity matrix, which can be used by different clustering techniques, such as the hierarchal, the k-mean, and the equivalence classes clustering algorithms.