A Causal Inference Method for Correcting Popularity Bias in Recommender Systems

A recommendation system and reasoning method technology, applied in the field of causal reasoning to correct the popularity deviation of recommendation systems, can solve problems such as difficult tuning and highly sensitive weighting strategies, reduce the time and difficulty of adjusting parameters, and improve recommendation performance.

Active Publication Date: 2022-07-15
UNIV OF SCI & TECH OF CHINA
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

While this method is theoretically sound, its weighting strategy is highly sensitive, making it difficult to tune

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Causal Inference Method for Correcting Popularity Bias in Recommender Systems
  • A Causal Inference Method for Correcting Popularity Bias in Recommender Systems
  • A Causal Inference Method for Correcting Popularity Bias in Recommender Systems

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

[0017] The embodiment of the present invention discusses the issue of popularity bias from a novel and basic perspective—causality, and proposes a causal reasoning method for correcting the popularity bias of a recommendation system, which is a model-independent counterfactual reasoning Methodological Framework (MACR). It is found that popularity bias exists in the direct effect from item nodes to ranking scores, so the intrinsic properties of items are t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a causal reasoning method for correcting the popularity deviation of a recommendation system, comprising: obtaining a matching score between a user and an item in a current recommendation system; predicting the item score according to the popularity of the item, and predicting the user score according to the user's preference; Aggregate the matching scores of users and items, item scores and user scores, predict the matching scores of users and items, and then remove the influence of popularity deviation to obtain the final matching scores of users and items. The method provided by the invention is a model-independent counterfactual reasoning framework, which can be applied to various recommendation systems. By eliminating the popularity deviation and improving the recommendation performance of the recommendation system, it can provide users with more high-quality and accurate personalized recommendation content .

Description

technical field [0001] The invention relates to the technical field of personalized recommendation, in particular to a causal reasoning method for correcting the popularity deviation of a recommendation system. Background technique [0002] Personalized recommendations have revolutionized many online applications, such as e-commerce, search engines, and conversational systems. A large number of recommendation models have been developed where the default optimization choice is to reconstruct the historical user-item interactions. However, the frequency distribution of items in interactive data is never balanced, and it is affected by exposure mechanisms, word-of-mouth effects, sales activities, product quality, and many other factors. In most cases, the frequency distribution of items is long-tailed, i.e. a few popular items have the majority of interactions in the dataset. This makes the classic training paradigm biased towards recommending popular items without revealing ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/9535G06N3/04G06N3/08G06N5/04
CPCG06F16/9535G06N5/04G06N3/08G06N3/045
Inventor 何向南魏天心冯福利陈佳伟易津锋
Owner UNIV OF SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap