Graph based Resource Recommender System

As technology improves day by day communication and browsing has become very much easier. However lots of information floods over the internet every moment thereby confusing the user as to select what or make decisions. To assist the user in identifying their interests and provide suggestions, recommendation systems came into existence. These systems filter the necessary content from large volumes of data to predict resources to the user. Common techniques used to implement recommender systems are content based approach or collaborative approach. However there are few limitations like data not being available for new users, ratings very sparse for resources. A graph based recommender system is proposed that makes useful recommendations by exploiting the significant content available. Clustering technique is used to identify the neighbourhood of the current user so that relevant resources are suggested. A weight based approach is used to calculate the ratings for the resources. This method is adopted to make the system less prone to data sparsity problem. This system is a web based client side application which makes recommendations by constructing user-resource graph and ranking the resources by a new method designed similar to that of search algorithms.