Implementasi Contextual Multi-Armed Bandit dengan Algoritma LinUCB untuk Rekomendasi Template pada Sistem Kampanye WhatsApp Berbasis Web
-
Reigan Chenartha
*
Universitas Pembangunan Jaya, Banten,
Indonesia
- Denny Ganjar Purnama Universitas Pembangunan Jaya, Banten, Indonesia (*) Corresponding Author
Abstract
Uniform WhatsApp campaign templates may overlook differences in timing and audience characteristics. This study develops an adaptive and auditable template recommendation module for a web-based campaign system. Design Science Research Methodology guides problem identification, objective definition, artifact development, demonstration, and evaluation. The disjoint Linear Upper Confidence Bound (LinUCB) algorithm treats Informative, Persuasive, and Urgency templates as three available actions. A 19-dimensional context vector encodes day, time bucket, audience, and a bias term. Template selection maximizes a linear reward estimate plus an exploration bonus, while the selected action is updated using delivered, read, and replied rewards weighted at 0.2, 0.3, and 0.5. Synthetic evaluation comprises 30 trials of 100 contexts each, with an exploration parameter of 0.3 and parent seed 1790543727863. LinUCB achieves a mean total reward of 39.520 with a standard deviation of 0.631, compared with 36.800 and 1.790 for the context-blind baseline. The mean improvement is 7.39%, with a mean regret of 2.769. The contribution is an integrated context–decision–feedback workflow with decision logging, operator control, and repeated evaluation isolated from operational data. These findings support artifact feasibility in the synthetic environment; improvements in real recipient responses require field validation.
Downloads
References
Antunes, P., Thuan, N. H., & Johnstone, D. (2022). Nature and purpose of visual artifacts in design science research. Information Systems and e-Business Management, 20(3), 515–550. https://doi.org/10.1007/s10257-022-00559-2
Blanchet, J., Xu, R., & Zhou, Z. (2024). Delay-adaptive learning in generalized linear contextual bandits. Mathematics of Operations Research, 49(1), 326–345. https://doi.org/10.1287/moor.2023.1358
Cai, R., Lu, R., Chen, W., & Hao, Z. (2024). Counterfactual contextual bandit for recommendation under delayed feedback. Neural Computing and Applications, 36(23), 14599–14613. https://doi.org/10.1007/s00521-024-09800-0
Cao, Z., Liang, Z., Wu, B., Zhang, S., Li, H., Wen, O., Rong, Y., & Zhao, P. (2023). Privacy matters: Vertical federated linear contextual bandits for privacy protected recommendation. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 154–166). Association for Computing Machinery. https://doi.org/10.1145/3580305.3599475
Chen, Q. (2024). A survey on contextual multi-armed bandits. Applied and Computational Engineering, 53(1), 287–295. https://doi.org/10.54254/2755-2721/53/20241593
Deffayet, R., Thonet, T., Hwang, D., Lehoux, V., Renders, J.-M., & de Rijke, M. (2024). SARDINE: Simulator for automated recommendation in dynamic and interactive environments. ACM Transactions on Recommender Systems, 2(3), Article 25. https://doi.org/10.1145/3656481
Gan, M., & Kwon, O.-C. (2022). A knowledge-enhanced contextual bandit approach for personalized recommendation in dynamic domains. Knowledge-Based Systems, 251, Article 109158. https://doi.org/10.1016/j.knosys.2022.109158
Ghoorchian, S., Kortukov, E., & Maghsudi, S. (2024). Non-stationary linear bandits with dimensionality reduction for large-scale recommender systems. IEEE Open Journal of Signal Processing, 5, 548–558. https://doi.org/10.1109/OJSP.2024.3386490
Gigli, M., & Stella, F. (2025). Multi-armed bandits for performance marketing. International Journal of Data Science and Analytics, 20(1), 151–165. https://doi.org/10.1007/s41060-023-00493-7
Haryanti, T., Rakhmawati, N. A., Subriadi, A. P., & Tjahyanto, A. (2022). The design science research methodology (DSRM) for self-assessing digital transformation maturity index in Indonesia. In 2022 IEEE 7th International Conference on Information Technology and Digital Applications (ICITDA) (pp. 1–7). IEEE. https://doi.org/10.1109/ICITDA55840.2022.9971171
Kayanda, A., Busagala, L., Oyelere, S., & Tedre, M. (2023). The use of design science and agile methodologies for improved information systems in the Tanzanian higher education context. The Electronic Journal of Information Systems in Developing Countries, 89(1), Article e12241. https://doi.org/10.1002/isd2.12241
Lattimore, T., & Szepesvári, C. (2020). Bandit algorithms. Cambridge University Press. https://doi.org/10.1017/9781108571401
Mardiana, N., & Latifah, N. (2025). Penguatan branding dan promosi produk UMKM Blitar melalui WhatsApp Business. Jurnal SOLMA, 14(3), 3398–3405. https://doi.org/10.22236/solma.v14i3.20638
Maulidan, M., Andayani Komara, M., & Kurniawan, I. (2023). Analisis kesuksesan aplikasi WhatsApp Business terhadap pelaku usaha UMKM dengan menggunakan pendekatan human organization technology fit (HOT FIT) model (Studi kasus: UMKM di Kabupaten Purwakarta). Jurnal Informatika Teknologi dan Sains (Jinteks), 5(3), 477–482. https://doi.org/10.51401/jinteks.v5i3.3305
Pires, P. R., Azevedo, G. F., Campos, P. L. P., Sereicikas, R. T., & Almeida, T. A. (2025). Exploitation over exploration: Unmasking the bias in linear bandit recommender offline evaluation. In Proceedings of the Nineteenth ACM Conference on Recommender Systems (pp. 736–745). Association for Computing Machinery. https://doi.org/10.1145/3705328.3748166
Qassimi, S., & Rakrak, S. (2025). Multi-objective contextual bandits in recommendation systems for smart tourism. Scientific Reports, 15(1), Article 13669. https://doi.org/10.1038/s41598-025-89920-2
Semenov, A., Rysz, M., Pandey, G., & Xu, G. (2022). Diversity in news recommendations using contextual bandits. Expert Systems with Applications, 195, Article 116478. https://doi.org/10.1016/j.eswa.2021.116478
Shi, Q., Xiao, F., Pickard, D., Chen, I., & Chen, L. (2023). Deep neural network with LinUCB: A contextual bandit approach for personalized recommendation. In Companion Proceedings of the ACM Web Conference 2023 (pp. 778–782). Association for Computing Machinery. https://doi.org/10.1145/3543873.3587684
Silva, N., Werneck, H., Silva, T., Pereira, A. C. M., & Rocha, L. (2022). Multi-armed bandits in recommendation systems: A survey of the state-of-the-art and future directions. Expert Systems with Applications, 197, Article 116669. https://doi.org/10.1016/j.eswa.2022.116669
Slivkins, A. (2019). Introduction to multi-armed bandits. Foundations and Trends in Machine Learning, 12(1-2), 1–286. https://doi.org/10.1561/2200000068
Taufiq, M. F., Ton, J.-F., Cornish, R., Teh, Y. W., & Doucet, A. (2022). Conformal off-policy prediction in contextual bandits. Advances in Neural Information Processing Systems, 35, 31512–31524. https://doi.org/10.52202/068431-2285
Thomas, M. A., Li, Y., & Lee, A. S. (2022). Generalizing the information systems artifact. Information Systems Research, 33(4), 1452–1466. https://doi.org/10.1287/isre.2022.1106
Zhao, Y., & Yang, L. (2024). Constrained contextual bandit algorithm for limited-budget recommendation system. Engineering Applications of Artificial Intelligence, 128, Article 107558. https://doi.org/10.1016/j.engappai.2023.107558
Zhu, Y., Foster, D. J., Langford, J., & Mineiro, P. (2022). Contextual bandits with large action spaces: Made practical. Proceedings of the 39th International Conference on Machine Learning, 162, 27428–27453. https://proceedings.mlr.press/v162/zhu22b.html
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Implementasi Contextual Multi-Armed Bandit dengan Algoritma LinUCB untuk Rekomendasi Template pada Sistem Kampanye WhatsApp Berbasis Web
Copyright (c) 2026 Reigan Chenartha, Denny Ganjar Purnama
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
