Genetic Algorithm and Statistical Markov Model Fusion for Predicting User's Surfing Behavior Using Sequence Patten Mining

mining useful for researchers in web prediction of the web browsing page sequence’s pattern analysis. The knowledge of navigated web page history of user browsing is helpful to predict the future set of page sequences that are likely to be visited by the user ahead of time. There is a wide scope for researchers to build and design prediction model based on browsing page sequences in sequence pattern mining. The main objective of prediction models is to achieve the better prediction accuracy and reduce the user latency. To achieve this objective, it is proposed to build the fusion of statistical Markov model and genetic algorithm based approach to improve the prediction accuracy. In addition, the genetic algorithm based approach is used to ease the modeling complexity of proposed system by generating optimal sequences of browsing patterns by reducing the size of search space. The proposed system is tested on the standard benchmark data sets to analyze prediction accuracy. The results outperformed by achieving 4% to 7% improvement over generalized Markov model.