A decision fusion of user page and concept matrices for enhancing next page prediction

With the advances in communication and technologies, the World Wide Web is becoming an important and rich source for information. The amount and variety of information available makes customization and personalized recommendations of utter importance. In this paper, we present a framework for the next page prediction that exploits users' access history combined with his semantic interests to generate personalized and accurate recommendations. The proposed framework offered a 54.3 % improvement in prediction accuracy over conventional methods for next page prediction. The suggested framework also employs user clustering to focus the search which reduced the prediction time by 63.4%.