Providing a web recommender system using markov chains and registration files structure

Today, due to the increasing growth of the internet and the huge amount of information we need systems to be able to recommend the most appropriate services and products to the user. The systems that do this are recommender systems. These systems are intelligently using artificial intelligence techniques to identify the interests of your users on the internet and suggest tailored offers to the user’s preferences and interests. Today, Markov models commonly used to predict web pages. For this purpose, in this research, we use a new Markov model and use the structure of registration files to predict the next pages obtained by the user. The proposed Markov model is based on a matrix of 1 to k and in the form of a Markov model which predicts the next pages. In order to reduce the complexity of the search space, as well as to better navigate to the recommender system, we use k-means clustering to group users. The results of the evaluations of the proposed method on the NASA web server log file and in the F-Measure criterion shows 0.57% superiority over BCF.