Nowadays the determination of the dynamics of sequential data, such as marketing, finance,social sciences or web research has receives much attention from researchers and scholars.Clustering of such data by nature is always a more challenging task. This paper investigates theapplications of different Markov models in web mining and improves a developed method forclustering web users, using hidden Markov models. In the first step, the categorical sequencesare transformed into a probabilistic space by hidden Markov model. Then, in the second step,hierarchical clustering, the performance of clustering process is evaluated with variousdistances criteria. Furthermore this paper shows implementation of the proposed improvementswith symmetric distance measure as Total-Variance and Mahalanobis compared with theprevious use of the proposed method (such as Kullback–Leibler) on the well-known Microsoftdataset with website user search patterns is more clearly result in separate clusters.
展开▼