卷积记忆图协同过滤

doi:10.13190/j.jbupt.2020-226

北京邮电大学学报 ›› 2021, Vol. 44 ›› Issue (3): 21-26.doi: 10.13190/j.jbupt.2020-226

卷积记忆图协同过滤

刘国桢, 陈鸿龙

中国石油大学(华东) 控制科学与工程学院, 青岛 266580

收稿日期:2020-11-03 出版日期:2021-06-28 发布日期:2021-06-23
通讯作者: 陈鸿龙(1984-),男,副教授,博士生导师,E-mail:chenhl@upc.edu.cn. E-mail:chenhl@upc.edu.cn
作者简介:刘国桢(1995-),男,硕士生.
基金资助:
国家自然科学基金项目（61772551）；国家自然科学基金国际合作与交流项目（62111530052）；中石油重大科技项目（ZD2019-183-003）

Convolutional Memory Graph Collaborative Filtering

LIU Guo-zhen, CHEN Hong-long

College of Control Science and Engineering, China University of Petroleum(East China), Qingdao 266580, China

Received:2020-11-03 Online:2021-06-28 Published:2021-06-23

摘要/Abstract

摘要： 针对推荐系统中用户和项目的向量表示问题，提出了一种端到端的具有记忆单元的图神经网络.在图神经网络中引入门控循环单元解决高阶连通节点间信息损失问题，可以使得用户和项目节点从高阶邻居获得更加完整的特征信息，然后利用卷积神经网络对网络输出层间的特征向量进行融合以获得不同阶段下用户的偏好.实验结果表明，与最优对比算法相比，采用所提卷积记忆图协同过滤推荐算法在4个数据集上的评分预测性能分别提升了1.98%，4.17%，9.27%和2.70%.

关键词: 图神经网络, 门控循环单元, 卷积神经网络, 评分预测, 推荐系统

Abstract: An end-to-end graph neural networks with memory unit is proposed for user vector representations and items in recommender systems. Gated recurrent unit is introduced to reduce the information loss between high-order connected nodes. This enables users and items nodes to obtain more complete feature information from high-order neighbor nodes. The convolutional neural networks are used to fuse feature vectors between different output layers to obtain users' preferences at different stages. Experiments on 4 datasets show that compared with the optimal comparison algorithms, the performance of proposed algorithm achieves gain of 1.98%, 4.17%, 9.27% and 2.7%, respectively.

Key words: graph neural networks, gated recurrent unit, convolutional neural networks, rating prediction, recommender systems

中图分类号:

TP301

刘国桢, 陈鸿龙. 卷积记忆图协同过滤[J]. 北京邮电大学学报, 2021, 44(3): 21-26.

LIU Guo-zhen, CHEN Hong-long. Convolutional Memory Graph Collaborative Filtering[J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(3): 21-26.

参考文献

[1] Srifi M, Oussous A, Lahcen A A, et al. Recommender systems based on collaborative filtering using review texts:a survey[J]. Information, 2020, 11(6):317.
[2] Wang Xiang, He Xiangnan, Wang Meng, et al. Neural graph collaborative filtering[C]//Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM, 2019:165-174.
[3] Xue Hongjian, Dai Xinyu, Zhang Jianbing, et al. Deep matrix factorization models for recommender systems[C]//International Joint Conference on Artificial Intelligence. Stockholm:IJCAI, 2017:3203-3209.
[4] Li Zekun, Cui Zeyu, Wu Shu, et al. Fi-GNN:modeling feature interactions via graph neural networks for CTR prediction[C]//International Conference on Information and Knowledge Management. New York:ACM, 2019:539-548.
[5] Chen Cheng, Meng Xiangwu, Xu Zhenghua, et al. Location-aware personalized news recommendation with deep semantic analysis[J]. IEEE Access, 2017, 5(99):1624-1638.
[6] He Xiangnan, Liao Lizi, Zhang Hanwang, et al. Neural collaborative filtering[C]//The Web Conference. New York:ACM, 2017:173-182.
[7] He Xiangnan, Chua T S. Neural factorization machines for sparse predictive analytics[C]//Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM, 2017:355-364.
[8] He Xiangnan, Du Xiaoyu, Wang Xiang, et al. Outer product-based neural collaborative filtering[C]//International Joint Conference on Artificial Intelligence. Stockholm:IJCAI, 2018:2227-2233.
[9] Xin Xin, Chen Bo, He Xiangnan, et al. CFM:convolutional factorization machines for context-aware recommendation[C]//International Joint Conference on Artificial Intelligence. Stockholm:IJCAI, 2019:3926-3932.
[10] Lamb C L, Garcez S A, Gori M, et al. Graph neural networks meet neural-symbolic computing:a survey and perspective[C]//International Joint Conference on Artificial Intelligence. Stockholm:IJCAI, 2020:4877-4884.
[11] Zhang Yufeng, Yu Xueli, Cui Zeyu, et al. Every document owns its structure:inductive text classification via graph neural networks[C]//Annual Meetings of Associa-tion for Computational Linguistics. Stroudsburg:ACL, 2020:334-339.
[12] Qi Xiaojuan, Liao Renjie, Jia Jiaya, et al. 3D graph neural networks for RGBD semantic segmentation[C]//IEEE International Conference on Computer Vision. New York:IEEE, 2017:5209-5218.
[13] Girshick B R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Computer Vision and Pattern Recognition. New York:IEEE, 2014:580-587.
[14] Guo Wei, Zhang Can, Guo Huifeng, et al. Multi-branch convolutional network for context-aware recommendation[C]//Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM, 2020:1709-1712.

[1]	刘美琴姬厚国白羽姚超赵耀. 基于雅可比动态逼近的图像去噪网络算法[J]. 北京邮电大学学报, 2024, 47(4): 63-70.
[2]	张光耀高欣张云凯刘婧叶平. 融合序列间显隐式关联的图卷积多维时序异常检测方法[J]. 北京邮电大学学报, 2024, 47(3): 75-82.
[3]	杜茂康彭俊杰胡勇进肖玲. 满足差分隐私的逻辑回归矩阵分解推荐算法[J]. 北京邮电大学学报, 2023, 46(3): 115-120.
[4]	吴进, 赵博, 汶恒, 王宇. 一种异构架构的图神经网络加速器优化研究一种异构架构的图神经网络加速器优化研究[J]. 北京邮电大学学报, 2023, 46(3): 19-24.
[5]	余佳润王亚峰. 一种低信噪比下基于深度学习的DoA估计方法[J]. 北京邮电大学学报, 2022, 45(6): 119-125.
[6]	叶康张淑军郭淇李辉崔雪红. 基于CM-Transformer的连续手语识别[J]. 北京邮电大学学报, 2022, 45(5): 49-53,78.
[7]	武兴坤罗涛刘爱军杨明张文静李剑峰李文秀. 面向主动脉缩窄辅助诊断的神经网络模型设计[J]. 北京邮电大学学报, 2022, 45(4): 1-6.
[8]	武兴坤, 罗涛, 刘爱军, 杨明, 张文静. 面向主动脉缩窄辅助诊断的神经网络模型设计[J]. 北京邮电大学学报, 2022, 45(4): 1-6.
[9]	许同乐, 孟良, 孔晓佳, 苏元浩, 孙砚飞. 基于EEMD的ICNN故障诊断方法[J]. 北京邮电大学学报, 2022, 45(2): 110-116.
[10]	李晓辉, 王先文, 樊韬, 刘佳文, 万宏杰. 基于卷积神经网络的高性能5G下行同步算法[J]. 北京邮电大学学报, 2022, 45(2): 117-123.
[11]	王靖, 姜竹青, 门爱东, 郭晓强, 王智康. 阶梯式图像去噪方法[J]. 北京邮电大学学报, 2022, 45(1): 52-57.
[12]	刘永利, 朱亚孟, 晁浩. 多策略MRFO算法的卷积神经网络超参数优化[J]. 北京邮电大学学报, 2021, 44(6): 83-88,95.
[13]	普运伟, 郭江, 刘涛涛, 吴海潇. 基于多学习单元卷积神经网络的雷达辐射源信号识别[J]. 北京邮电大学学报, 2021, 44(6): 74-82.
[14]	王艺霏, 莫爽, 吴文睿, 范少华, 肖丁. 基于内外卷积网络的网络入侵检测[J]. 北京邮电大学学报, 2021, 44(5): 94-100.
[15]	冀振燕, 吴梦丹, 杨春, 李俊东. 可扩展的融合多源异构数据的推荐模型[J]. 北京邮电大学学报, 2021, 44(3): 106-111.

卷积记忆图协同过滤

Convolutional Memory Graph Collaborative Filtering

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价