Recently keen interests shown in studying the goal behind a user’s Web query, so that this goal can be used to improve the quality of a search engine’s results in turn improves the popularity of web pages. Advertisement fees can be decided based on this factor. Personalization is now becoming common term for improving E-commerce services and attract more users. Today’s recommender system provides suggestion for specific items but drawback that service provider can increase the ratings of specific product and unnecessarily popularity increases. Traditional data protection mechanisms focus on access control and secure transmission, which provide security only against malicious third parties, but not the service provider. This leads to misguiding the users while purchasing some products, so privacy is violated. Many different approaches including web usage mining have been applied to the basic problem of developing accurate and efficient recommendation systems. Online business transactions and the success of E-commerce depend greatly on the effective design of a product recommender mechanism. Our proposal is founded on homomorphic encryption, which is used to obscure the private rating information of the customers from the service provider. While the user’s privacy is respected by the service provider, by generating recommendations using encrypted customer ratings, the service provider’s commercially valuable item–item similarities are protected against curious entities, in turn. Agent based Web mining has advantages of both Web mining and Agent: it can mine data efficiently and intelligently, so it is becoming more and more important in modern E-business.