Abstract: Recommender systems are applied in a variety of applications like movies, music, research articles, social tags and restaurants etc. As these systems have become extremely common in recent years. Providing the good recommendations to the customer based on their usage patterns is the major focus of this paper. Prior research has reported that desired property of recommendation algorithm is the stability and has important implications on userís trust and acceptance of recommendation. Two scalable, general purpose meta algorithmic approaches based on bagging and iterative smoothing that can be used in conjunction with different traditional recommendation algorithms to improve their stability are introduced in .The proposed system is not only for improving the recommendation stability, but are actually able to provide good recommendations based on userís usage patterns.
Keywords: Bagging, Collaborative Filtering, Iterative Smoothing, Recommender Systems, Recommendation Stability.