A Recommender System (RS) is the most significant technologies that handle the information overload problem of Retrieval Information by suggesting users with correct and related items. Today, abundant recommender systems have been developed for different fields and we put an effort on collaborative filtering (CF) recommender system. There are several problems in the recommender system such as Cold Start, Synonymy, Shilling Attacks, Privacy, Limited Content Analysis and Overspecialization, Grey Sheep, Sparsity, Scalability and Latency Problem. The current research explored the privacy in CF recommender system and defined the perspective privacy attributes (user's identity, password, address, and postcode/location) which are required to be addressed. Using the base models as Homomorphic and Hash Encryption scheme, we have proposed a hybrid model Homomorphic Hash Encryption (H2E) model that addressed the privacy issues according to defined objectives in the current study. Furthermore, in order to evaluate the privacy level, H2E was implementing in medicine recommender system and compared the consequences with existing state-of-the-art privacy protection mechanisms. It was observed that H2E outperform to other models with respect to determined privacy objectives. Leading to user's privacy, H2E can be considered a promising model for CF recommender systems.