Using Clustering and Markov Model in Predicting Web User's Next Request

Different techniques have been proposed for reducing user perceived latency. Web prefetching is one of techniques aimed at reducing this latency. This technique uses a prediction algorithm for predicting user's future requests. In this paper, a method is proposed for predicting user's next requests. This method can be applied for web prefetching. In the proposed method, a model which is a combination of Markov model and clustering is proposed for prediction. Time features of accesses in the website are also considered in the proposed model. The result of experiments shows the improvement of proposed model with respect to the Markov model.