基于场因子分解的xDeepFM推荐模型

doi:10.3969/j.issn.0255-8297.2024.03.012

应用科学学报 ›› 2024, Vol. 42 ›› Issue (3): 513-524.doi: 10.3969/j.issn.0255-8297.2024.03.012

基于场因子分解的xDeepFM推荐模型

李子杰^1,2, 张姝^1,2, 欧阳昭相^1,3, 王俊^1,2, 吴迪^1,2

1. 云南师范大学民族教育信息化教育部重点实验室, 云南昆明 650500;
2. 云南师范大学云南省智慧教育重点实验室, 云南昆明 650500;
3. 德宏师范高等专科学校信息学院, 云南芒市 678400

收稿日期:2021-12-15 发布日期:2024-06-06
通信作者: 张姝，副教授，研究方向为智能知识处理。E-mail: zhangshu@ynnu.edu.cn E-mail:zhangshu@ynnu.edu.cn
基金资助:
国家自然科学基金（No. 62266054, No. 62166050）、云南省教育厅科学研究基金（No. 2023Y0535）资助

xDeepFM Recommendation Model Based on Field Factorization

LI Zijie^1,2, ZHANG Shu^1,2, OUYANG Zhaoxiang^1,3, WANG Jun^1,2, WU Di^1,2

1. Key Laboratory of Educational Information for Nationalities, Ministry of Education, Yunnan Normal University, Kunming 650500, Yunnan, China;
2. Yunnan Key Laboratory of Smart Education, Yunnan Normal University, Kunming 650500, Yunnan, China;
3. School of Information, Dehong Teacher's College, Mangshi 678400, Yunnan, China

Received:2021-12-15 Published:2024-06-06

摘要/Abstract

摘要： 极深因子分解机（eXtreme deep factorization machine,xDeepFM）是一种基于上下文感知的推荐模型，它提出了一种压缩交叉网络对特征进行阶数可控的特征交叉，并将该网络与深度神经网络进行结合以优化推荐效果。为了进一步提升xDeepFM在推荐场景下的表现，提出一种基于场因子分解的xDeepFM改进模型。该模型通过场信息增强了特征的表达能力，并建立了多个交叉压缩网络以学习高阶组合特征。最后分析了用户场、项目场设定的合理性，并在3个不同规模的MovieLens系列数据集上通过受试者工作特征曲线下面积、对数似然损失指标进行性能评估，验证了该改进模型的有效性。

关键词: 推荐算法, 极深因子分解机, 场因子分解, 深度学习

Abstract: The eXtreme deep factorization machine (xDeepFM) is a context-aware recommendation model integrating a compressed interaction network for controllable feature cross-ordering. The network is combined with deep neural network to optimize the recommendation performance. To further improve xDeepFM’s performance in recommended scenarios, eXtreme deep field factorization machine (xDeepFFM) is proposed in this paper. The improved model enhances feature expression capabilities through field information and uses multiple compressed interaction networks to learn higher-order combinatorial features based on field information. Furthermore, this paper analyzes the rationality of the setting of user field and item field. The effectiveness of the improved model is evaluated using area under curve and Log-likelihood loss metrics on three public datasets of different sizes.

Key words: recommendation algorithm, eXtreme deep factorization machine (xDeepFM), field factorization, deep learning

中图分类号:

TP391

李子杰, 张姝, 欧阳昭相, 王俊, 吴迪. 基于场因子分解的xDeepFM推荐模型[J]. 应用科学学报, 2024, 42(3): 513-524.

LI Zijie, ZHANG Shu, OUYANG Zhaoxiang, WANG Jun, WU Di. xDeepFM Recommendation Model Based on Field Factorization[J]. Journal of Applied Sciences, 2024, 42(3): 513-524.

参考文献

[1] Goldberg D, Nichols D, Oki B M, et al. Using collaborative filtering to weave an information tapestry [J]. Communications of the ACM, 1992, 35(12): 61-70.
[2] Koren Y, Bell R, Volinsky C. Matrix factorization techniques for recommender systems [J]. Computer, 2009, 42(8): 30-37.
[3] Sedhain S, Menon A K, Sanner S, et al. Autorec: autoencoders meet collaborative filtering [C]//The 24th International Conference on World Wide Web, 2015: 111-112.
[4] Shan Y, Hoens T R, Jiao J, et al. Deep crossing: web-scale modeling without manually crafted combinatorial features [C]//The 22nd ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2016: 255-262.
[5] He X N, Liao L Z, Zhang H W, et al. Neural collaborative filtering [C]//The 26th International Conference on World Wide Web, 2017: 173-182.
[6] Qu Y R, Cai H, Ren K, et al. Product-based neural networks for user response prediction [C]//IEEE 16th International Conference on Data Mining (ICDM), 2016: 1149-1154.
[7] Cheng H T, Koc L, Harmsen J, et al. Wide & deep learning for recommender systems [C]//The 1st Workshop on Deep Learning for Recommender Systems, 2016: 7-10.
[8] Guo H F, Tang R M, Ye Y M, et al. DeepFM: a factorization-machine based neural network for CTR prediction [DB/OL]. 2017[2021-12-15]. http: //arxiv.org/abs/1703.04247.
[9] Guo Q Y, Zhuang F Z, Qin C, et al. A survey on knowledge graph-based recommender systems [J]. IEEE Transactions on Knowledge and Data Engineering, 2022, 34(8): 3549-3568.
[10] Sun R, Cao X Z, Zhao Y, et al. Multi-modal knowledge graphs for recommender systems [C]//The 29th ACM International Conference on Information & Knowledge Management, 2020: 1405-1414.
[11] Zhao W X, Mu S L, Hou Y P, et al. RecBole: towards a unified, comprehensive and efficient framework for recommendation algorithms [DB/OL]. 2020[2021-12-15]. https://arxiv.org/abs/2011.01731.
[12] Zhou K, Wang H, Zhao W X, et al. S3-rec: self-supervised learning for sequential recommendation with mutual information maximization [C]//The 29th ACM International Conference on Information & Knowledge Management, 2020: 1893-1902.
[13] Zhou G, Mou N, Fan Y, et al. Deep interest evolution network for click-through rate prediction [DB/OL]. 2020[2021-12-15]. https://arxiv.org/abs/1809.03672.
[14] 陈卓, 李倩, 杜军威. 面向化工领域社区问答的答案质量预测研究[J]. 东北师大学报(自然科学版), 2021, 53(3): 81-88. Chen Z, Li Q, Du J W. Prediction of answer quality for community Q & A in chemical industry [J]. Journal of Northeast Normal University (Natural Science Edition), 2021, 53(3): 81-88. (in Chinese)
[15] Zhang J, Wu Z C, Li F, et al. Deep attentional factorization machines learning approach for driving safety risk prediction [J]. Journal of Physics: Conference Series, 2021, 1732: 012007.
[16] Lian J X, Zhou X H, Zhang F Z, et al. xDeepFM: combining explicit and implicit feature interactions for recommender systems [C]//The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018: 1754-1763.
[17] Juan Y, Zhuang Y, Chin W S, et al. Field-aware factorization machines for CTR prediction [C]//The 10th ACM Conference on Recommender Systems, 2016: 43-50.
[18] 黄若然, 崔莉, 韩传奇. 推荐系统中稀疏情景预测的特征—类别交互因子分解机[J]. 计算机研究与发展, 2022, 59(7): 1553-1568. Huang R R, Cui L, Han C Q. Feature-over-field interaction factorization machine for sparse contextualized prediction in recommender systems [J]. Journal of Computer Research and Development, 2022, 59(7): 1553-1568. (in Chinese)
[19] Zhang W N, Du T M, Wang J. Deep learning over multi-field categorical data [C]//European Conference on Information Retrieval, 2016: 45-57.
[20] He X N, Chua T S. Neural factorization machines for sparse predictive analytics [C]//The 40th International ACM SIGIR Conference on Research and Development in Information Retrieval, 2017: 355-364.
[21] Harper F M, Konstan J A. The MovieLens datasets: history and context [J]. ACM Transactions on Interactive Intelligent Systems, 2015, 5(4): 1-19.
[22] Ziegler C N, Mcnee S M, Konstan J A, et al. Improving recommendation lists through topic diversification [C]//The 14th International Conference on World Wide Web, 2005: 22-32.

基于场因子分解的xDeepFM推荐模型

xDeepFM Recommendation Model Based on Field Factorization

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

[1]	沈坤烨, 周晓飞, 费晓波, 陈雨中, 张继勇, 颜成钢. 基于边缘感知深度残差网络的带钢表面缺陷检测[J]. 应用科学学报, 2023, 41(6): 978-988.
[2]	黄显培, 孟庆祥. 基于多特征卷积神经网络的哨兵二号影像地物分类[J]. 应用科学学报, 2023, 41(5): 766-776.
[3]	张春森, 朱江乐, 张学芬, 刘旭东, 史书. 融合自注意力机制与生成对抗网络的DEM空洞填充[J]. 应用科学学报, 2023, 41(5): 789-800.
[4]	魏洁玲, 马秀丽, 金彦亮, 王瑞. 基于VPN通道下的加密流量分类算法[J]. 应用科学学报, 2023, 41(4): 646-656.
[5]	萧晓彤, 丁建伟, 张琪. 基于局部和全局梯度上升的分段后门防御[J]. 应用科学学报, 2023, 41(2): 218-227.
[6]	王婷, 王娜, 崔运鹏, 刘娟. 基于深度学习的医疗电子数据特征学习方法[J]. 应用科学学报, 2023, 41(1): 41-54.
[7]	曾静, 李莹, 戚小莎, 吉根林. 多层记忆增强生成对抗网络二次预测的视频异常检测方法[J]. 应用科学学报, 2023, 41(1): 80-94.
[8]	姜晓勇, 李忠义, 黄朗月, 彭孟乐, 徐书杨. 神经网络剪枝技术研究综述[J]. 应用科学学报, 2022, 40(5): 838-849.
[9]	罗长银, 陈学斌, 宋尚文, 张淑芬, 刘之瑜. 基于深度学习的联邦集成算法[J]. 应用科学学报, 2022, 40(3): 493-510.
[10]	雷前慧, 潘丽丽, 邵伟志, 胡海鹏, 黄瑶. 基于三重注意力机制的新冠肺炎病灶分割模型[J]. 应用科学学报, 2022, 40(1): 105-115.
[11]	朱丽, 杨青, 吴涛, 李晨, 李铭. 基于CNN和Bi-LSTM的脑电波情感分析[J]. 应用科学学报, 2022, 40(1): 1-12.
[12]	欧巧凤, 肖佳兵, 谢群群, 熊邦书. 基于深度学习的车检图像多目标检测与识别[J]. 应用科学学报, 2021, 39(6): 939-951.
[13]	康慧娴, 易标, 吴汉舟. 文本隐写及隐写分析综述[J]. 应用科学学报, 2021, 39(6): 923-938.
[14]	郭毓博, 陆军, 段鹏启. 基于深度学习的竹笛吹奏技巧自动分类[J]. 应用科学学报, 2021, 39(4): 685-694.
[15]	李文举, 何茂贤, 张耀星, 陈慧玲, 李培刚. 基于卷积神经网络和投票机制的轨道板裂缝检测[J]. 应用科学学报, 2021, 39(4): 627-640.