Social recommendation method based on oblivious transfer

Abstract

The present invention discloses a social recommendation method based on oblivious transfer, which comprises the following steps that (1) an SN terminal and an RBS terminal share data distribution conditions of user historic behavior recording matrixes to each other, and the two terminals only calculate items of sim(u, u<j>) is not equal to 0 and w(u<j>, i) is not equal to 0 (only the data distribution conditions are shared by the two terminals, and data value information of the two terminals are not involved); (2) the two terminals respectively complete recommendation score calculation of articles by utilization of the OT multiplication protocol, and the RBS and the SN cannot presume data of the other one only according to the owned calculation scores; (3) the recommendation scores of the articles are secretly shared by the RBS and the SN in an addition sum manner; and (4) a recommendation result is obtained by comparison. Through the mode, the social recommendation method based on the oblivious transfer is advantaged by extremely little calculation cost, extremely strong adaptability and extremely high efficiency, and promotion of the social recommendation method based on the oblivious transfer has wide market prospect.

Description

technical field [0001] The invention relates to the field of secure multi-party computing, in particular to a social recommendation method based on inadvertent transmission. Background technique [0002] Neighborhood-based algorithms are the most basic algorithms in recommender systems, which have not only been deeply studied in academia, but also widely used in the industry. Neighborhood-based algorithms are divided into two categories: one is user-based collaborative filtering algorithm, and the other is item-based collaborative filtering algorithm. [0003] The collaborative filtering method mainly predicts the user's future behavior based on the user's historical behavior and interests, so a large amount of user historical behavior data has become an important component and prerequisite of the recommendation system. For many websites in the initial stage, without the accumulation of a large amount of user data, the collaborative filtering method cannot play a good role....

AI Technical Summary

Problems solved by technology

[0006] The main technical problem to be solved by the present invention is to provide a social recommendation method based on inadvertent transmission, through inadvertent transmission and cooperation to complete personalized social recommendation without disclosing the private data of any participant, based on friend relationship Recommendation can increase the trust of recommendation, and the use of social networks can effectively solve the cold start problem of traditional recommendation methods. It has high security and strong adaptability, and has broad market prospects in the popularization of social recommendation methods based on inadvertent transmission.

Images