Logo PTI Logo FedCSIS

Proceedings of the 21st Conference on Computer Science and Intelligence Systems (FedCSIS)

Annals of Computer Science and Information Systems, Volume 47

Scalable and Interpretable Agentic AI Recommendation Framework for Big Data Online Product Recommendation

,

DOI: http://dx.doi.org/10.15439/2026F7287

Citation: Shu-Hsien Liao,

Full text

Abstract. Deep learning recommenders suffer from black-box opacity and struggle with heterogeneous, incomplete multi‑criteria data. Current Agentic AI systems lack symbolic reasoning for verifiable decisions. We propose a Scalable Agentic Recommendation Framework (ARF) integrating Rough Set Theory (RST) with Analytic Hierarchy Process (AHP) and Association Rule Mining (ARM). Our event‑driven agent architecture embeds symbolic rules for real‑time reasoning and autonomous adaptation. A two‑stage algorithm processes nominal, ordinal, and ratio data without forced discretization. Validated via consumer simulations, our approach reduces inference latency by 86\% and rule‑matching cost by 95\% while maintaining interpretability. This bridges symbolic AI and agentic systems for big data e‑commerce.

References

  1. Z. Gu, F. Liu, and J. Chen, A Proactive Agent Collaborative Framework for Zero-Shot Multimodal Medical Reasoning. Advanced Intelligent Systems, vol 7 no.8, Article 2400840, 2025. https://doi.org/10.1002/aisy.202400840
  2. Z. Jing, X. Xu, and Y. Jin, Emotion vs. information: Understanding the effect of AI-powered call systems. Decision Support Systems, vol.201, 114579, 2026. https://doi.org/10.1016/j.dss.2025.114579
  3. Z. Elgammal, M. Taleb Albrijawi, and R. Alhajj, Digital twins in healthcare: a review of AI-powered practical applications across health domains. Journal of Big Data. 2025; 12:234. https://doi.org/10.1186/s40537-025-01280-w
  4. S. Greco, J. Figueira, and M. Ehrgott, Multiple criteria decision analysis. Springer, 2016.
  5. X. Geng, Z. Yang, and Z. Ma, Association rule-based classification: A comprehensive review. Expert Systems with Applications, vol.280 Article 127454, 2025. https://10.1016/j.eswa.2025.127454
  6. M. Bukhari, M. Maqsood, and F. Adil, An actor-critic based recommender system with context-aware user modeling. Artificial Intelligence Review, vol.58: Article 138 2025. https://doi.org/10.1007/s10462-025-11134-9
  7. M.Z. Naser, A review of machine learning with small and limited data. Journal of Big Data, vol.13 Article 28, 2025. https://doi.org/10.1186/s40537-025-01346-9
  8. G. Göksel, A. Aydın, and C. Kaleli, A novel missing value imputation for multi-criteria recommender systems. Information Sciences, vol. 712 Article 122139, 2025. https://doi.org/10.1016/j.ins.2025.122139
  9. A. Xavier, R. Fragoso, and M.B. Costa Freitas, Building sustainability composite indicators using a multi-criteria approach. European Journal of Operational Research, vol.326 no.2, pp.326-342, 2025. https://10.1016/j.ejor.2025.04.024
  10. H. Li, H. Fan, and J. Wang, Data-driven heterogeneous ensemble forecasting framework incorporating multi-scale criteria. Expert Systems with Applications, vol. 302 Article 130537, 2026. 10.1016/j.eswa.2025.130537
  11. Z. Pawlak, Rough sets. International Journal of Computer and Information Sciences, vol. 11 no.5, pp.341-356, 1982. http://dx.doi.org/10.1007/BF01001956
  12. Z. Pawlak, Rough classification. International Journal Human-Computer Studies, vol. 51 no.2, pp.369-383, 1999. https://doi.org/10.1016/S0020-7373(84)80022-X
  13. W. Xu, Y. Yan, and X. Li, Sequential rough set: a conservative extension of Pawlak's classical rough set. Artificial Intelligence Review, vol.58 Article 9. 2025. https://doi.org/10.1007/s10462-024-10976-z
  14. D. Feng, W. Liu, and S. Peng, CoRoFR: Community Detection of Feature Graph Improves Feature Selection Using Robust Fuzzy Rough Set. Advanced Intelligent Systems, vol.7 no.12, Article e202500044, 2025. https://doi.org/10.1002/aisy.202500044
  15. C. Zheng, M. Shareduwan, and M. Kasihmuddin, Dual-phase feature selection using adaptive neighborhood rough sets. Engineering Applications of Artificial Intelligence, vol. 160 Article 111899, 2025. https://doi.org/10.1016/j.engappai.2025.111899
  16. W. Qiu, C. Liu, and D. Miao, Attribute reduction based on adaptive neighborhood rough sets. Expert Systems with Applications, vol.271 Article 126618, 2025. https:// 10.1016/j.eswa.2025.126618
  17. R. Agrawal, and T. Imieliński, A. Swami, Mining association rules between sets of items in large databases. ACM SIGMOD Record, vol. 22 no.2, Article 207-216, 1993. https://doi.org/10.1145/170036.17007
  18. S. Aljehani, and Y. Alotaibi, Preserving privacy in association rule mining using multi-threshold particle swarm optimization. Information Sciences, vol. 692 Article 121673, 2025. https://doi.org/10.1016/j.ins.2024.121673
  19. X. Geng, J. Zhang, and Z. Ma, Evidential association rule learning for semi-supervised activity recognition. Information Sciences, vol.728 Article 122811, 2026. https://doi.org/10.1016/j.ins.2025.122811
  20. X. Lu, and J. Wu, Combining association-rule-guided sequence augmentation with listwise contrastive learning. Information Processing & Management, vol. 62 no.3, Article 103999, 2025. 10.1016/j.ipm.2024.103999
  21. T.Sharma, and F. Emmert-Streib, Deep mining the textual gold in relation extraction. Artificial Intelligence Review, vol.58 Article 34, 2025. https://doi.org/10.1007/s10462-024-11042-4
  22. X. Chu, B. Sun, and N. Chen, Multi-modal and multi-criteria conflict analysis model based on deep learning and dominance-based rough sets. Information Fusion, vol.113 Article 102636, 2025. 10.1016/j.inffus.2024.102636