A Hybrid Approach to Improving Scalability in Collaborative Filtering

The process of filtering information or patterns using techniques involving collaboration among multiple agents or data sources is known as collaborative filtering [14]. Applications of collaborative filtering typically involve very large data sets. Techniques of Collaborative filtering have been applied to many different fields such as sensing and monitoring data in mineral exploration, environmental sensing over large areas, financial data, such as financial service institutions or in web applications where the focus is on user information. It is based on the concept that everything is related to everything else [9]. One such popular field of development of collaborative filtering is Recommendation Systems. Recommendation systems were developed to guide users in a personalized way to a large set of possible options matching their choices and requirements. A content-based recommender system matches the attributes of a user‘s preferences and interests to the attributes of an object (item). On the other hand a collaborative filtering takes the approach of matching one user‘s choices to the choices of another user. The basic assumption behind this method is that other users‘ opinions can be selected and aggregated in such a way as to provide a reasonable prediction of the active user‘s preference. Hence a new hybrid and scalable recommendation system has been proposed in this research that combines techniques from Content-Based Recommender Systems, Collaborative Filtering, Location Aware Recommender Systems and Spatial Autocorrelation. General Terms Collective intelligence, Similarity search, Collaborative search engine , Content discovery platform, Decision support system.