User preference commodity recommendation method based on meta-learning

Abstract

The invention discloses a user preference commodity recommendation method based on meta-learning. According to the method, the feature precision can be improved by fusing auxiliary information. The user channel is established through meta-learning to realize user feature conversion, the problem of user interest attenuation caused by over-targeted personalized recommendation of a single service is solved, the diversity and novelty of a recommendation system are improved, meanwhile, the accuracy of target domain user commodity recommendation can be improved, and the user experience is improved. According to the method, through the re-learning ability of meta-learning, if a new task arrives, the mapping relation can be quickly established, so that the new task can be quickly learned on the basis of obtaining existing knowledge, and recommendation of the target domain is realized.

Description

technical field [0001] The invention belongs to the field of cross-domain recommendation, and in particular relates to a method for recommending user preference commodities based on meta-learning. Background technique [0002] Since the 21st century, with the rapid development of information technology and computer technology, business activities have also been internationalized, virtualized and paperless. The traditional consumer industry is gradually moving towards e-commerce. At the same time, with the continuous development and improvement of the domestic e-commerce environment, online platform shopping has become a part of people's lives. People have more and more strict requirements for personalized recommendation of platform products. The cold start problem of commercial platforms is extra important, and cross-domain recommendation can solve the problem of user interest attenuation caused by overly targeted personalized recommendation of a single business, and improve...

AI Technical Summary

Problems solved by technology

This model uses entities in different domains to build a heterogeneous graph shared by all domains. By integrating all interaction information and content information into a shared graph, the characteristics of users and items in different domains can be fully considered, making it possible for users and The modeling of items is more accurate, but the cost of model construction is too high and depends on a large amount of auxiliary information

Images