融合注意力与深度因子分解机的时间上下文推荐模型

摘要/Abstract

摘要： 对于许多在线电商，预测用户购买商品的可能性至关重要。由于用户与商品的交互通常是高维且稀疏的，所以深度因子分解机算法（DeepFM）将因子分解机算法（FM）与深度神经网络（DNN）结合在一起，用FM处理低阶特征组合，用DNN处理高阶特征组合，通过并行的方式组合这2种方法，很好地解决了高维稀疏的问题。但是，它忽略了用户购买商品的先后性问题，也就是时间上下文信息。针对这一缺陷，本文提出一种融合注意力（Attention）与DeepFM的时间上下文推荐模型（DeepAFM），更好地利用用户与商品交互的时间上下文信息，相比较于未加入时间上下文信息的DeepFM模型，AUC提升了1.84%。对比验证结果表明，DeepAFM模型具有更优越的性能。

关键词: 推荐系统, 时间上下文, 注意力机制, 深度因子分解机

Abstract: For many online E-commerce companies, it is very important to predict the possibility of consumers purchasing goods. Because the interaction between users and commodities is usually high-dimensional and sparse, the deep factor decomposer algorithm (DeepFM) combines the factor decomposer algorithm (FM) with the deep neural network (DNN), uses FM to deal with low-order feature combination, uses DNN to deal with high-order feature combination, and combines the two methods in parallel, which solves the problem of high-dimensional sparse. However, it ignores the order of purchase, that is, time context information. Aiming at this defect, this paper proposes a time context recommendation system (DeepAFM) which integrates attention and DeepFM, which makes better use of the time context information of user and commodity interaction. Compared with DeepFM model without time context information, the AUC is increased by 1.84%. The results of the comparison and verification show that DeepAFM model has better performance.

Key words: recommendation system, temporal context, attention mechanism, DeepFM

刘亦欣, 王家伟, 李自力. 融合注意力与深度因子分解机的时间上下文推荐模型[J]. 计算机与现代化, 2021, 0(11): 22-27.

LIU Yi-xin, WANG Jia-wei, LI Zi-li. Temporal Context Recommendation Model Integrating Attention and DeepFM[J]. Computer and Modernization, 2021, 0(11): 22-27.

参考文献［25］

［1］	BATMAZ Z, YUREKLI A, BILGE A, et al. A review on deep learning for recommender systems: Challenges and remedies［J］. Artificial Intelligence Review, 2019,52(1):1-37.
［2］	ADOMAVICIUS G, TUZHILIN A. Context-aware recommender systems［M］// Recommender Systems Handbook. Springer, Boston, MA, 2011:217-253.
［3］	〖JP+1〗BALABANOVIC M, SHOHAM Y. Fab: Content-based, collaborative recommendation［J］. Communications of the ACM, 1997,40(3):66-72.
［4］	BURKE R. Hybrid recommender systems: Survey and experiments［J］. User Modeling and User-adapted Interaction, 2002,12(4):331-370.
［5］	RENDLE S. Factorization machines［C］// The 10th IEEE International Conference on Data Mining. 2010:995-1000.
［6］	JUAN Y, ZHUANG Y, CHIN W S, et al. Field-aware factorization machines for CTR prediction［C］// 2016 ACM Conference on Recommender Systems. 2016:43-50.
［7］	GUO H F, TANG R M, YE Y M, et al. DeepFM: A factorization-machine based neural network for CTR prediction［C］//Proceedings of the 26th International Joint Conference on Artificial Intelligence. 2017:1725-1731.
［8］	窦羚源,王新华,孙克. 融合标签特征和时间上下文的协同过滤推荐算法［J］. 小型微型计算机系统, 2016,37(1):48-52.
［9］	黄文明,卫万成,张健,等. 基于注意力机制与评论文本深度模型的推荐方法［J］. 计算机工程, 2019,45(9):176-182.
［10］	郁豹,李振华,张凯,等. 基于DeepFM模型的广告推荐系统研究［J］. 计算机应用与软件, 2019,36(7):307-310.
［11］	黎丹雨. 基于多特征融合的电影推荐系统［J］. 计算机与现代化, 2019(8):121-126.
［12］	谢浩然,卫巍,杨志辉,等. 基于TF-IDF的新型广播电视节目协同推荐流程［J］. 计算机与现代化, 2019(9):65-71.
［13］	吴韦俊,李烨. 神经因子分解机推荐模型改进研究［J］.软件导刊, 2020,19(4):115-118.〖HJ1.83mm〗
［14］	崔鑫. 基于异构信息网络的推荐系统［J］. 计算机与现代化, 2020(12):13-19.
［15］	汤小月,周康,王凯. 一种空间上下文感知的提及目标推荐方法［J］. 软件学报, 2020,31(4):1189-1211.
［16］	陈劲松,孟祥武,纪威宇,等. 基于多维上下文感知图嵌入模型的兴趣点推荐［J］. 软件学报, 2020,31(12):3700-3715.
［17］	HIDASI B, KARATZOGLOU A, BALTRUNAS L, et al. Session-based Recommendations with Recurrent Neural Networks［J/OL］. (2015-11-21)［2016-03-29］. https://arxiv.org/abs/1511.06939.
［18］	LIU B, WEI Y, ZHANG Y, et al. Transferable contextual bandit for cross-domain recommendation［C］// Proceedings of the 2018 AAAI Conference on Artificial Intelligence. 2018:3619-3626.
［19］	BENDADA W, SALHA G, BONTEMPELLI T. Carousel personalization in music streaming APPs with contextual bandits［C］// The 14th ACM Conference on Recommender Systems. 2020:420-425.
［20］	MA Y F, NARAYANASWAMY B, LIN H B, et al. Temporal-contextual recommendation in real-time［C］// Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2020:2291-2299.
［21］	HANSEN C, HANSEN C, MAYSTRE L, et al. Contextual and sequential user embeddings for large-scale music recommendation［C］// The 14th ACM Conference on Recommender Systems. 2020:53-62.
［22］	WU S, TANG Y Y, ZHU Y Q, et al. Session-based recommendation with graph neural networks［C］// Proceedings of the AAAI Conference on Artificial Intelligence. 2019,33:346-353.
［23］	SANTANA M R O, MELO L C, CAMARGO F H F, et al. Contextual meta-bandit for recommender systems selection［C］// The 14th ACM Conference on Recommender Systems. 2020:444-449.
［24］	JRV P. Predictability limits in session-based next item recommendation［C］// Proceedings of the 13th ACM Conference on Recommender Systems. 2019:146-150.
［25］	HE X, AN B, LI Y H, et al. Contextual user browsing bandits for large-scale online mobile recommendation［C］// The 14th ACM Conference on Recommender Systems. 2020:63-72.

[1]	王秋忆, 周浩, 郑婷婷. 改进RetinaNet的电力设备目标检测方法[J]. 计算机与现代化, 2024, 0(01): 47-52.
[2]	林启钊, 彭志平, 郭棉, 崔得龙. 基于双向多步预测的炉管温度场重构方法[J]. 计算机与现代化, 2024, 0(01): 53-58.
[3]	李亚平, 王军防, 余红梅, 窦一民, 肖媛, 田继林. Regformer：基于稀疏注意力的输油管道水力压降预测方法[J]. 计算机与现代化, 2024, 0(01): 59-66.
[4]	邱凯星, 冯广. 基于双重特征注意力的多标签图像分类模型[J]. 计算机与现代化, 2023, 0(12): 41-47.
[5]	张浩洋, 尹梓名, 乐珺怡, 沈达聪, 束翌俊, 杨自逸, 孔祥勇, 龚伟. 3D-SPRNet: 一种基于并行解码器和双注意力机制的胆囊癌分割模型[J]. 计算机与现代化, 2023, 0(12): 59-66.
[6]	张伯泉, 麦海鹏, 陈嘉敏, 逄锦聚. 基于高灰度值注意力机制的脑白质高信号分割[J]. 计算机与现代化, 2023, 0(12): 67-75.
[7]	宁娟, 周庆华, 曾小为. 改进YOLOv7算法在西林瓶轧盖缺陷检测中的应用[J]. 计算机与现代化, 2023, 0(12): 82-86.
[8]	王宇航, 董宝良, 公超, 尚真真, 姚康宁. 基于意图识别的空中群目标动态威胁评估[J]. 计算机与现代化, 2023, 0(12): 100-104.
[9]	罗明杰, 冯开平. 基于沙漏结构与注意力机制的轻量级人脸表情识别方法[J]. 计算机与现代化, 2023, 0(11): 89-94.
[10]	张嘉琪, 徐啟蕾. 基于NAM-YOLO网络的苹果缺陷检测算法[J]. 计算机与现代化, 2023, 0(10): 53-58.
[11]	叶思佳, 魏延, 杜韩宇, 邓金枝. 结合注意力机制的HRNet图像语义分割算法[J]. 计算机与现代化, 2023, 0(10): 65-69.
[12]	陈嘉敏, 张伯泉, 麦海鹏. 基于特征融合的海马体分割[J]. 计算机与现代化, 2023, 0(08): 1-6.
[13]	刘续, 查可可. 一种用于机场特种车辆作业的环境目标识别方法[J]. 计算机与现代化, 2023, 0(08): 18-24.
[14]	王鸿, 葛红. 基于注意力机制和语义相似度的跨模态哈希检索[J]. 计算机与现代化, 2023, 0(08): 44-53.
[15]	欧阳飞, 吴旭, 向东升. 基于改进YOLOX的垃圾分类检测方法[J]. 计算机与现代化, 2023, 0(08): 68-73.