SLGP Header

Enhancing Recommender System for Matrimonial Sites using Collaborative Filtering Method

IJCSEC Front Page

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.
One of the key problems in today’s world is to deal with lots of information. The information is available to the user whether or not he looks for it. This growth of information on the internet led to the development of recommender systems. The Recommender System (RS) attempts to reduce this information overload and helps individuals to make decisions. The main goal is to recommend items based on the preferences of the end user. Here ‘item’ is a general term that refers to what the system recommends to the user. The main task of the RS is to predict items of user interest and then recommend them. The popular existing recommender systems are for e-shopping, MovieLens for movies etc. Many different types of RS have been developed in recent years. Most of the existing RS uses the Collaborative Filtering approach.
Collaborative Filtering:
The collaborative filtering [1] makes automatic predictions (filtering) about the interest of individual by collecting preference information from many users (collaborating). There are model-based collaborative filtering and memory-based collaborative filtering. The Model-based CF makes predictions on real data by learning a model. It then makes recommendations by computing the expected value of user prediction based on his ratings on other items. Memory-based CF uses user rating data to compute similarity between users and then recommendations are given. This system first identifies a set of users called neighbours. Then it uses different algorithms to combine preferences of neighbours to provide recommendation to the active user. It is also known as user-based collaborative filtering and is most widely used in practice.


  1. B. Sarwar, G. Karypis, J. Konstan, and J. Riedl, “Item-based collaborative filtering recommendation algorithms,” in Proc. 10th Int. Conf. WWW, 2001, pp. 285–295.
  2. L. A. S. Pizzato, T. Rej, T. Chung, I. Koprinska, and J. Kay, "Recon: a reciprocal recommender for online dating," in RecSys, pp. 207-214, 2010.
  3. Smita Khavate, Mrs. M. A. Potey“Partial fuzzy recommender system for Indian matrimony”.
  4. Qi Liu, Enhong Chen, Chris H. Q. Ding Enhancing “Collaborative Filtering by User Interest Expansion via Personalized Ranking”
  5. Francesco Ricci · Lior Rokach · Bracha Shapira ·Paul B. Kantor “Recommender Systems Handbook” pg 107 – 140.
  6. Michal Holub, Maria Bielikova “Estimation of User Interest in Visited Web Page”
  7. Feng Qiu, Junghoo Cho ”Automatic Identification of User Interest For Personalized Search”
  8. Tamil Matrimony .com
  9. S. Debnath, N. Ganguly, and P.Mitra, “Feature weighting in content based recommendation system using social network analysis,” in Proc. 17th Int. Conf. WWW, 2008, pp. 1041–1042.
  10. G. Adomavicius and A. Tuzhilin, "Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions," IEEE Trans. Knowl. Data Eng., vol. 17, no. 6, pp. 734-749, 2005.
  11. L. Brozovsky and V. Petricek, "Recommender system for online dating service," in Proceedings of Znalosti 2007 Conference, Ostrava), VSB, 2007.
  12. R. Nayak, M. Zhang, and L. Chen, "A social matching system for an online dating network: A preliminary study," in Proceedings of the 2010 IEEE International Conference on Data Mining Workshops, ICDMW '10, (Washington, DC, USA), pp. 352-357, IEEE Computer Society, 2010.
  13. B. Pang and L. Lee, “Opinion Mining and Sentiment Analysis,” Foundationsand Trends in Information Retrieval, vol. 2, nos. 1–2, 2008, pp. 1–135.