A Novel User Click Behavior Identification Method Based on Hidden Semi-Markov Model

The accurate user click behavior identification is the precondition of web user behavior characters researching, but the growing complexity of web page makes it difficult to be identified. In this paper, a novel method to identify user click behavior is proposed and implemented. First, the sequence of user HTTP requests is described through hidden semi-Markov model, which is then computed into user state sequence by parameter estimation algorithm and state estimation algorithm. Finally, the user click behavior can be identified according to the state transition in the user state sequence. Numerical results based on real campus network traffic data are presented to demonstrate the effectiveness of the user click behavior method.