A multi-domain content recommendation method and system that fuses novelty

By integrating matrix factorization based on novelty and calculation of user feature similarity, the user item rating matrix is ​​optimized, solving the problems of new users and data sparsity in the recommendation system, improving the novelty and accuracy of recommendations, and achieving recommendation results that are closer to user preferences.

CN116257697BActive Publication Date: 2026-03-24HAINAN UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202310229536.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-03-10
Publication Date
2026-03-24
Estimated Expiration
2043-03-10

AI Technical Summary

Technical Problem

Existing recommendation systems have not yet effectively solved problems such as new users, data sparsity, and insufficient novelty of recommendation results, especially the cold start problem and data sparsity problem.

Method used

A system filtering method that integrates novelty matrix factorization and user features is adopted. The user item rating matrix is ​​optimized by BiasSVD model, a novelty factor is added, and features such as user gender, age and occupation are combined. The loss function is optimized by gradient descent and the parameters are updated using gradient descent. User feature similarity and Pearson similarity are calculated to predict the rating of unrated items.

Benefits of technology

It improves the novelty and accuracy of recommendations, mitigates the negative impact of data sparsity on recommendation performance, enhances the accuracy of the recommendation system, solves the cold start problem, and makes the recommendation results closer to the user's true preferences.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application provides a multi-field content recommendation method and system with novelty fusion, and the application optimizes a user-item rating matrix based on a BiasSVD model, adds a novelty factor to predict the user rating matrix, and alleviates problems such as unsatisfactory recommendation effect caused by data sparseness in a traditional model; similarity is calculated based on user feature information, cold start problems caused by excessive dependence of a traditional recommendation algorithm on historical data are solved, the recommendation result is closer to the real rating, and the recommendation accuracy is improved; the novelty relationship between users is introduced, and the novelty of the recommendation result of a traditional collaborative filtering recommendation algorithm is improved. The method disclosed in the application can alleviate the problem of sparse rating data, and can improve the novelty of the recommendation result, so that the recommendation result is not too single, thereby avoiding user aesthetic fatigue.
Need to check novelty before this filing date? Find Prior Art