Federal learning recommendation system and method, storage medium and electronic equipment

Abstract

The invention relates to a federal learning recommendation system and method, a storage medium and electronic equipment. The federal learning recommendation method comprises the steps that a user score matrix is mapped to a confused user score matrix through a confusion function, and each element of the user score matrix indicates whether a plurality of users score a plurality of commodities or not through binary values; and generating a federal learning recommendation result by using the confused user score matrix, wherein the federal learning recommendation result is associated with the user score matrix. Wherein the confusion function is obtained through optimization meeting a preset constraint condition, and the preset constraint condition is based on elements which have the same position but different binary values between the user score matrix and the confused user score matrix. In this way, user privacy protection is enhanced, and meanwhile good overall calculation efficiency and resource occupancy rate are achieved.

Description

technical field [0001] The present application relates to the technical field of machine learning, especially to the technical field of data security and privacy protection, and specifically to a federated learning recommendation system, method, storage medium and electronic equipment. Background technique [0002] With the development of machine learning technology, recommender system (Recommender System, RS) and related information filtering, personalized recommendation, collaborative filtering and other technologies have made great progress, and are widely used in e-commerce, news, film and television, catering, financial services, etc. It has been widely used in various industries. The basic principle of RS is to send the user's score (it can also refer to information that can reflect the user's interest or demand, such as preferences, scores, choices, etc.) to the recommendation server, and then the recommendation server uses these user scores to provide personalized se...

AI Technical Summary

Problems solved by technology

However, it is difficult for existing technologies to determine the balance between ensuring user privacy and avoiding excessive time resource occupation.

Images