基于图神经网络的推荐算法综述  

A review of recommendation algorithms based on graph neural networks

作　　者：黄海新[1] 杜亭亭 HUANG Haixin;DU Tingting(Shenyang Ligong University,Shenyang 110000,China)

机构地区：[1]沈阳理工大学,辽宁沈阳110000

出　　处：《通信与信息技术》2023年第3期65-69,共5页Communication & Information Technology

摘　　要：互联网的快速发展带来了大量信息,用户很难在有限时间内获取所需信息。基于图神经网络的推荐系统通过为用户提供个性化推荐,挖掘适应当前时代背景的潜在兴趣偏好来解决这一难题。通过回顾传统推荐算法的分类,对基于深度学习的推荐算法提供了一种新的分类方式。最后,根据使用的信息类型和推荐任务提出一种基于GNN的推荐分类。With the rapid development of the Internet,it is difficult for users to get the information they need in a limited time.The recommendation system based on graph neural network solves this problem by providing personalized recommendation for users.By reviewing the classification of traditional recommendation algorithms,it provides a new classification method for recommendation al⁃gorithms based on deep learning.Finally,a GNN-based recommendation classification is proposed according to the information type and recommendation task used.

关 键 词：推荐算法 图神经网络 图表示学习 深度学习 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]