magazinelogo

Advances in Computer and Communication

ISSN Online: 2767-2875 CODEN: ACCDC3
Frequency: quarterly Email: acc@hillpublisher.com
Total View: 899112 Downloads: 178306 Citations: 149 (From Dimensions)

Volumes & Issues

Current Issue

All Volumes

ArticleOpen Access http://dx.doi.org/10.26855/acc.2025.04.004

TrAdaBoostR2-based Domain Adaptation for Generalizable Revenue Prediction in Online Advertising Across Various Data Distributions

Yu Qiao1, Kaixian Xu2, Zhaoyan Zhang3, Alan Wilson4,*

1Meta Platforms, Inc., Bellevue, WA 98005, USA.

2Risk & Quant Analytics, BlackRock, Jersey City, NJ 07097, USA.

3Zhongke Zhidao (Beijing) Technology Co., Ltd., Beijing 102627, China.

4Intact Financial Corporation, Toronto, ON M5G 0A1, Canada.

*Corresponding author: Alan Wilson

Published: May 13,2025

Abstract

Accurately predicting advertising revenue is critical for optimizing budget allocation and campaign performance in the online advertising industry. However, domain shifts caused by differences in audience behavior, campaign settings, or temporal factors pose significant challenges to model generalization. In this study, we propose a TrAdaBoostR2-based domain adaptation framework to enhance the generalizability of revenue prediction models across varying data distributions. We first simulate domain discrepancies by applying K-Means clustering to partition a real-world advertising dataset into distinct source and target domains. Then, several classical regression models—including Decision Tree, Random Forest, Gradient Boosting, and K-Nearest Neighbors—are integrated into a TrAdaBoostR2 pipeline to perform adaptive learning across domains. Experimental results show that models enhanced with domain adaptation significantly outperform their non-adapted counterparts, achieving lower MAE and RMSE and higher R² scores. The TrAdaBoost + Random Forest model, in particular, achieved the best performance with an R² of 0.7796. This study highlights the effectiveness of boosting-based domain adaptation in addressing distributional heterogeneity in online advertising environments and offers a practical pathway for building robust and transferable predictive systems.

Keywords

Online advertising; revenue prediction; TrAdaBoostR2; domain adaptation

References

[1] Gan Y, Zhu D. The research on intelligent news advertisement recommendation algorithm based on prompt learning in end-to-end large language model architecture. Innov Appl Eng Technol. 2024 Aug 30:1-9.

[2] Gan Y, Ma J, Xu K. Enhanced e-commerce sales forecasting using EEMD-integrated LSTM deep learning model. J Comput Methods Eng Appl. 2023 Nov 11:1.

[3] Zeng A, Ruan W, Li Y. “Who” should be the center? Differentiated tourism digital advertisement design based on the tourist attraction attributes. Asia Pac J Tour Res. 2025 Apr 3:1-8.

[4] Würfel M, Han Q, Kaiser M. Online advertising revenue forecasting: an interpretable deep learning approach. In: 2021 IEEE International Conference on Big Data (Big Data); 2021 Dec 15; p. 1980-9.

[5] Li S, Chen Y, Chen Y, Huang Y. Predicting advertisement revenue of social-media-driven content websites: toward more efficient and sustainable social media posting. Sustainability. 2022 Apr 2;14(7):4225.

[6] Moon H, Lee T, Seo J, Park C, Eo S, Aiyanyo ID, et al. Return on advertising spend prediction with task decomposition-based LSTM model. Mathematics. 2022 May 11;10(10):1637.

[7] Xu K, Cai Y, Wilson A. Inception residual RNN-LSTM hybrid model for predicting pension coverage trends among private-sector workers in the USA, 2025.

[8] Xu J, Xu K, Wang Y, Shen Q, Li R. A k-means algorithm for financial market risk forecasting. arXiv:2405.13076 [Preprint]. 2024 May 21.

[9] Lu P, Zhang Z. The model of food nutrition feature modeling and personalized diet recommendation based on the integration of neural networks and K-means clustering. J Comput Biol Med. 2025 Feb 11;5(1).

[10] Zhang Z. RAG for personalized medicine: a framework for integrating patient data and pharmaceutical knowledge for treatment recommendations. Optim Appl Mach Learn. 2024 Dec 10;4(1).

[11] Wilson A, Ma J. MDD-based domain adaptation algorithm for improving the applicability of the artificial neural network in vehicle insurance claim fraud detection. Optim Appl Mach Learn. 2025 Feb 13;5(1).

[12] Huang W, Zhou T, Ma J, Chen X. An ensemble model based on fusion of multiple machine learning algorithms for remaining useful life prediction of lithium battery in electric vehicles. Innov Appl Eng Technol. 2025 Mar 4:1-2.

[13] Wiles O, Gowal S, Stimberg F, Alvise-Rebuffi S, Ktena I, Dvijotham K, et al. A fine-grained analysis on distribution shift. arXiv:2110.11328 [Preprint]. 2021 Oct 21.

[14] Cai T, Namkoong H, Yadlowsky S. Diagnosing model performance under distribution shift. arXiv:2303.02011 [Preprint]. 2023 Mar 3.

[15] Malinin A, Band N, Chesnokov G, Gal Y, Gales MJ, Noskov A, et al. Shifts: a dataset of real distributional shift across multiple large-scale tasks. arXiv:2107.07455 [Preprint]. 2021 Jul 15.

[16] Huang W, Ma J. Analysis of vehicle fault diagnosis model based on causal sequence-to-sequence in embedded systems. Optim Appl Mach Learn. 2023 Sep 12;3(1).

[17] Huang W, Cai Y, Zhang G. Battery degradation analysis through sparse ridge regression. Energy Syst. 2024 Dec 10;4(1).

[18] Ding X, Meng Y, Xiang L, Boden-Albala B. Stroke recurrence prediction using machine learning and segmented neural network risk factor aggregation. Discov Public Health. 2024 Oct 7;21(1):119.

[19] Zhu D, Gan Y, Chen X. Domain adaptation-based machine learning framework for customer churn prediction across varing distributions. J Comput Methods Eng Appl. 2021 Oct 21:1-4.

[20] Zhang G, Zhou T, Cai Y. CORAL-based domain adaptation algorithm for improving the applicability of machine learning models in detecting motor bearing failures. J Comput Methods Eng Appl. 2023 Nov 3:1-7.

[21] Chen X, Wang M, Zhang H. Machine learning-based fault prediction and diagnosis of brushless motors. Eng Adv. 2024 Aug 6;4(3).

[22] Zhang H, Zhu D, Gan Y, Xiong S. End-to-end learning-based study on the Mamba-ECANet model for data security intrusion detection. J Inf Technol Policy. 2024 Sep 7:1-7.

[23] Zhou Z, Wu J, Cao Z, She Z, Ma J, Zu X. On-demand trajectory prediction based on adaptive interaction car following model with decreasing tolerance. In: 2021 International Conference on Computers and Automation (CompAuto); 2021 Sep 7; p. 67-72.

[24] Ma J, Zhang Z, Xu K, Qiao Y. Improving the applicability of social media toxic comments prediction across diverse data platforms using residual self-attention-based LSTM combined with transfer learning. Optim Appl Mach Learn. 2022 Aug 10;2(1).

[25] Ma J, Chen X. Fingerprint image generation based on attention-based deep generative adversarial networks and its application in deep siamese matching model security validation. J Comput Methods Eng Appl. 2024 Dec 12:1-3.

[26] Birim S, Kazancoglu I, Mangla SK, Kahraman A, Kazancoglu Y. The derived demand for advertising expenses and implications on sustainability: a comparative study using deep learning and traditional machine learning methods. Ann Oper Res. 2024 Aug;339(1):131-61.

[27] Würfel M, Han Q, Kaiser M. Online advertising revenue forecasting: an interpretable deep learning approach. In: 2021 IEEE Inter-national Conference on Big Data (Big Data); 2021 Dec 15; p. 1980-9.

[28] Su H, Du Z, Li J, Zhu L, Lu K. Cross-domain adaptative learning for online advertisement customer lifetime value prediction. In: Proceedings of the AAAI Conference on Artificial Intelligence; 2023 Jun 26; vol. 37, no. 4; p. 4605-13.

[29] Pan S. A survey on transfer learning. IEEE Trans Knowl Data Eng. 2010;22(10):1345-59.

[30] Gopalan R, Li R, Chellappa R. Domain adaptation for object recognition: an unsupervised approach. In: 2011 International Conference on Computer Vision; 2011 Nov; p. 999-1006.

[31] Jhuo IH, Liu D, Lee DT, Chang SF. Robust visual domain adaptation with low-rank reconstruction. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition; 2012 Jun; p. 2168-75.

[32] Pan SJ, Tsang IW, Kwok JT, Yang Q. Domain adaptation via transfer component analysis. IEEE Trans Neural Netw. 2010;22(2):199-210.

[33] Ajakan H, Germain P, Larochelle H, Laviolette F, Marchand M. Domain-adversarial neural networks. arXiv:1412.4446 [Preprint]. 2014 Dec 14.

[34] Ganin Y, Lempitsky V. Unsupervised domain adaptation by backpropagation. In: International Conference on Machine Learning; 2015 Jun; p. 1180-9.

[35] Glorot X, Bordes A, Bengio Y. Domain adaptation for large-scale sentiment classification: a deep learning approach. In: Proceedings of the 28th International Conference on Machine Learning; 2011; p. 513-20.

[36] Hamerly G, Elkan C. Learning the k in k-means. Adv Neural Inf Process Syst. 2003;16.

[37] Ahmed M, Seraj R, Islam SM. The k-means algorithm: a comprehensive survey and performance evaluation. Electronics. 2020 Aug 12;9(8):1295.

[38] Zhu Y, Zhao Y, Song C, Wang Z. Evolving reliability assessment of systems using active learning-based surrogate modelling. Physica D. 2024 Jan 1;457:133957.

[39] Ye X, Luo K, Wang H, Zhao Y, Zhang J, Liu A. An advanced AI-based lightweight two-stage underwater structural damage detection model. Adv Eng Inform. 2024 Oct 1;62:102553.

[40] Wang Z, Zhao Y, Song C, Wang X, Li Y. A new interpretation on structural reliability updating with adaptive batch sampling-based subset simulation. Struct Multidiscip Optim. 2024 Jan;67(1):7.

[41] O’Donnell R, Saks M, Schramm O, Servedio RA. Every decision tree has an influential variable. In: 46th Annual IEEE Symposium on Foundations of Computer Science; 2005 Oct 23; p. 31-9.

[42] Biau G, Scornet E. A random forest guided tour. Test. 2016 Jun;25(2):197-227.

[43] Steinbach M, Tan PN. kNN: k-nearest neighbors. In: The top ten algorithms in data mining; 2009 Apr 9; p. 165-76.

[44] Zhang Z, Jung C. GBDT-MO: gradient-boosted decision trees for multiple outputs. IEEE Trans Neural Netw Learn Syst. 2020 Aug 4;32(7):3156-67.

[45] Cleeman J, Agrawala K, Malhotra R. Accelerated and inexpensive machine learning for manufacturing processes with incomplete mechanistic knowledge. Manuf Lett. 2023 Sep 1;37:53-6.

[46] Liu Y, Gu R, Gao X, Bai L. Transfer learning based programmable Raman amplifier for flexible multi-band optical network. In: 2023 Asia Communications and Photonics Conference/2023 International Photonics and Optoelectronics Meetings; 2023 Nov 4; p. 1-3.

[47] Wenjun D, Fatahizadeh M, Touchaei HG, Moayedi H, Foong LK. Application of six neural network-based solutions on bearing capacity of shallow footing on double-layer soils. Steel Compos Struct. 2023;49(2):231-44.

[48] Zhao Y, Dai W, Wang Z, Ragab AE. Application of computer simulation to model transient vibration responses of GPLs reinforced doubly curved concrete panel under instantaneous heating. Mater Today Commun. 2024 Mar 1;38:107949.

[49] Ma J, Wilson A. Mitigating FGSM-based white-box attacks using convolutional autoencoders for face recognition. Optim Appl Mach Learn. 2023 Dec 15;3(1).

How to cite this paper

TrAdaBoostR2-based Domain Adaptation for Generalizable Revenue Prediction in Online Advertising Across Various Data Distributions

How to cite this paper: Yu Qiao, Kaixian Xu, Zhaoyan Zhang, Alan Wilson. (2025) TrAdaBoostR2-based Domain Adaptation for Generalizable Revenue Prediction in Online Advertising Across Various Data Distributions. Advances in Computer and Communication6(2), 67-80.

DOI: http://dx.doi.org/10.26855/acc.2025.04.004