Recommendation system filters the information and shows only the data that are relevant to the user. In this paper, we have considered external factors in recommendation such as scrutability, algorithm accuracy and user interface as well as internal factors such as similarity, diversity, negative similarity, serendipity. The characteristics of each factor in similarity based recommendation system have been clearly examined and a model Fuzzy Inference System has been developed. Factors that influence in determining the recommendation are discussed and evaluated using rule based fuzzy inference model. © 2015 IEEE.