Enhancing Recommender System for Matrimonial Sites using Collaborative Filtering Method

Recommender Systems helps users to find items of interest from a large number of available items. Collaborative Filtering is the commonly used technology for recommender systems. The role of recommender system in matrimonial sites is profile matching based on the preferences given by the users. The users of matrimonial sites have a problem of overloaded choices of partners. This is because the currently used collaborative-filtering-based recommender systems focus only on the information about the way the users interact with the system and the original interests of the user is not identified. In this paper we provide a collaborative-filtering-based recommender system that identifies the user’s latent interest and provides top-n recommendations to the active user based on preferences and interests of the user. The top-n recommendations are identified by ranking recommended profiles in terms of weightage. This helps the users of matrimonial sites to easily identify a perfect match. Moreover opinion mining can be used to identify profiles which are of interest to maximum users. This explores the minds of current generation as to what kind of profiles the users are mostly interested in. Keywords-Recommender System, Collaborative Filtering, Opinion Mining.