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.