Recommender system with web usage mining based on fuzzy c means and neural networks

To provide consumers with a choice of using a web content recommender system that generates personalized recommendations in a timely manner based on a model of their own habits and behaviors'. Initially, data cleaning is performed and then semantically enhanced web usage logs are prepared. Then, web access activities are identified according to periodic and resource attributes. Fuzzy c means clustering algorithm has been applied and based on support and confidence values, periodic pattern are obtained crossing a particular threshold. Drawbacks are then overcome by using neural networks based recommender systems, in order to provide recommendations for every possible case of periodic and resource attributes.