Privacy protection method and system in collaborative filtering recommendation

The invention discloses a privacy protection method in collaborative filtering recommendation. The privacy protection method comprises the following steps: 1) a user sends a request to a recommendation system; 2) the recommendation system sends a signal collection instruction to a server, and the server transmits a query result to a recommendation server; and 3) the recommendation server analyzes the query result and sends a recommendation result to the user. According to the privacy protection method in collaborative filtering recommendation, after sparse user item scores are filled, noise obeying Gaussian distribution is randomly added, so that attackers can be prevented from breaking specific distribution and inversely solving a matrix, and part of privacy information is recovered from interfered data. And on the other hand, the influence of the time weight on the recommendation precision is considered, and the recommendation precision is improved, so that the user can enjoy high-quality recommendation ser