Prediction Method and System for Object Evaluation Value, Recommendation Method and System

By adding noise to user and object data and training aSDAE model, the problem of insufficient robustness and anti-interference in personalized recommendations is solved, and higher recommendation accuracy and robustness are achieved.

CN111695024BActive Publication Date: 2025-06-17BEIJING JINGDONG SHANGKE INFORMATION TECH CO LTD +1
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN201910181966.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-03-11
Publication Date
2025-06-17
Estimated Expiration
2039-03-11

Smart Images

  • Figure CN111695024B_ABST
    Figure CN111695024B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and system for predicting an object evaluation value, a recommendation method and system, an electronic device, and a storage medium. The prediction method includes: establishing a database; querying target user parameters, target object parameters, and object evaluation values from the database, and constructing a user feature vector, an object feature vector, and an object rating matrix; using the user feature vector, object feature vector, and object rating matrix processed by adding noise as training samples to train an aSDAE model to obtain a collaborative filtering model; the output parameters of the collaborative filtering model include a user latent factor vector and an object latent factor vector; the product of the two is used to predict the evaluation value of the user for the object. The present invention can accurately predict the evaluation value of the user for an unknown object, that is, accurately predict the preference degree of the user for the unknown object.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Data processing method and system, computer system and computer readable storage medium

    CN110610392A