文章详情

题名 · 作者 · 摘要 · 关键词 · DOI · 参考文献

基于多分支特征融合模型的广告点击率预测研究

王志格1，潘胜科2

（1.广州应用科技学院 计算机学院，广东 肇庆 526072；2.广东政达合顺项目管理有限公司，广东 河源 517000）

栏目：人工智能 · 第 2026 年第 20 期 · 页码：42-48,55 · 2096-4706（2026）20-0042-08 · 中图分类号：TP391.3；TP183 · 文献标识码：A · DOI：10.19850/j.cnki.2096-4706.2026.20.009

摘要

点击率（CTR）预测是在线广告与推荐系统的核心任务，其准确性直接影响用户体验与平台收益。针对现有模型在多模态特征交互建模、上下文动态感知及分支特征融合方面的不足，文章提出一种融合解纠缠注意力、上下文特征提取和场矩阵分解机的多分支特征融合CTR预估模型（简称MFNet）。模型多分支融合层耦合上下文动态感知、解纠缠注意力交互与场矩阵分解关联建模，实现特征表示的深度融合。在公开数据集Criteo和MovieLens-1M上的实验表明，所提模型与DESTINE模型相比AUC指标分别提升了0.07个百分点和0.44个百分点，交叉熵损失函数Logloss分别降低了0.71个百分点和0.13个百分点。

关键词

点击率预测；分支特征融合；解纠缠注意力；上下文特征提取；场矩阵分解机

参考文献

  1. [1] Zhang W N，Qin J R，Guo W，et al. Deep Learning for Click-Through Rate Estimation [C]//Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence Survey Track，2021：4695-4703.
  2. [2] Rendle S. Factorization machines [C]//2010 IEEE International Conference on Data Mining，2010：995-1000.
  3. [3] Pan J W，Xu J，Ruiz A L，et al. Field-weighted Factorization Machines for Click-Through Rate Prediction in Display Advertising [PP/OL].arXiv:1806.03514 [cs.LG].https://arxiv.org/abs/1806.03514.
  4. [4] Sun Y，Pan J W，Zhang A，et al. FM2: Field-matrixed Factorization Machines for Recommender Systems [PP/OL].arXiv:2102.12994 [cs.IR].https://arxiv.org/abs/2102.12994.
  5. [5] He X N，Chua T S. Neural Factorization Machines for Sparse Predictive Analytics [C]//Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval，2017：355-364.
  6. [6] Covington P，Adams J，Sargin E. Deep Neural Networks for YouTube Recommendations [C]//ACM Conference on Recommender Systems，2016：191-198.
  7. [7] Liu B，Tang R M，Chen Y Z，et al. Feature Generation by Convolutional Neural Network for Click-Through Rate Prediction [C]//WWW '19：The World Wide Web Conference，2019：1119-1129.
  8. [8] Guo H F，Chen B，Tang R M，et al. An Embedding Learning Framework for Numerical Features in CTR Prediction [C]//KDD'21: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining，2021：2910-2918.
  9. [9] Xiao J，Ye H，He X N，et al. Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networks [C]//Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence Main track，2017：3119-3125.
  10. [10] Vaswani A，Shazeer N，Parmar N，et al. Attention is All You Need [C]//NIPS'17: Proceedings of the 31st International Conference on Neural Information Processing Systems，2017：6000-6010.
  11. [11] Song W，Shi C，Xiao Z，et al. Autoint: Automatic Feature Interaction Learning via Self-attentive Neural Networks [C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management，2019：1161-1170.
  12. [12] Xu Y，Zhu Y，Yu F，et al. Disentangled Self-Attentive Neural Networks for Click-Through Rate Prediction [C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management，2021：3553-3557.
  13. [13] Leng J，Liu Y，Wang Z，et al. CrossNet: Detecting Objects as Crosses [J].IEEE Transactions on Multimedia，2022，24：861-875.
  14. [14] Cheng H-T，Koc L，Harmsen J，et al. Wide & Deep Learning for Recommender Systems [C]//DLRS 2016: Proceedings of the 1st Workshop on Deep Learning for Recommender Systems，2016：7-10.
  15. [15] Guo H-F，Tang R-M，Ye Y-M，et al. DeepFM: A Factorization-machine Based Neural Network for CTR Prediction [C]//IJCAI'17: Proceedings of the 26th International Joint Conference on Artificial Intelligence，2017：1725-1731.
  16. [16] Lian J X，Zhou X H，Zhang F Z，et al. xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems [PP/OL].arXiv:1803.05170 [cs.LG].https://arxiv.org/abs/1803.05170.
  17. [17] Yu Y T，Wang Z，Yuan B. An Input-aware Factorization Machine for Sparse Prediction [C]//Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence，2019：1466-1472.
  18. [18] Lu W T，Yu Y T，Chang Y Z，et al. A Dual Input-aware Factorization Machine for CTR Prediction [C]//Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI-20)，2020：3139-3145.
  19. [19] Huang T，Zhang Z，Zhang J. FiBiNET: Combining Feature Importance and Bilinear Feature Interaction for Click-through Rate Prediction [C]//Proceedings of the 13th ACM Conference on Recommender Systems，2019：169-177.
  20. [20] Hu J，Shen L，Albanie S，et al. Squeeze-and-Excitation Networks [C]//IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，42（8）：2011-2023.
  21. [21] Wang F Y，Wang Y X，Li D S，et al. Enhancing CTR Prediction with Context-Aware Feature Re-presentation Learning [C]//SIGIR'22: Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval，2022：343-352.
  22. [22] Li Z Y，Cheng W，Chen Y. Interpretable Click-Through Rate Prediction through Hierarchical Attention [C]//WSDM'20: Proceedings of the 13th International Conference on Web Search and Data Mining，2020：313-321.
  23. [23] Li Y J，Tarlow D，Brockschmidt M ，et al. Gated Graph Sequence Neural Networks [PP/OL].arXiv:1511.05493 [cs.LG].https://arxiv.org/abs/1511.05493?context = cs.AI.
  24. [24] Kipf T N，Welling M. Semi-supervised Classification with Graph Convolutional Networks [PP/OL].arXiv:1609.02907 [cs.LG].https://arxiv.org/abs/1609.02907.
  25. [25] Li Z，Cui Z，Wu S，et al. Fi-GNN: Modeling Feature Interactions via Graph Neural Networks for CTR Prediction [C]//The 28th ACM International Conference on Information and Knowledge Management，2019：539-548.
  26. [26] Qu Y R，Cai H，Ren K，et al. Product-Based Neural Networks for User Response Prediction[PP/OL].arXiv:1611.00144 [cs.LG].https://arxiv.org/abs/1611.00144?context = cs.
  27. [27] Zhao Z S，Yang S，Liu G H，et al. FINT: Field-aware INTeraction Neural Network For CTR Prediction [PP/OL].arXiv:2107.01999 [cs.IR].https://arxiv.org/abs/2107.01999.
  28. [28] Wang R，Shivanna R，Cheng D，et al. DCNV2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems [C]//Proceedings of the Web Conference 2021，2021：1785-1797.
  29. [29] Yan B C，Wang P J，Zhang K，et al. APG: Adaptive Parameter Generation Network for Click-Through Rate Prediction [J].Advances in Neural Information Processing Systems，2022，35：24740-24752.

基金项目

广州应用科技学院智能算法研究所课程群教学团队（2025TD005）

作者简介

王志格（1997—），男，汉族，安徽池州人，助教，硕士研究生，研究方向：点击率预估推荐、目标检测等；潘胜科（1992—），男，汉族，广东河源人，人工智能工程师，本科，研究方向：推荐系统。