Autonomous Agent-Based Swarm Intelligence for Adaptive Privacy-Preserving Data Mining.

Saved in:

Bibliographic Details
Title:	Autonomous Agent-Based Swarm Intelligence for Adaptive Privacy-Preserving Data Mining.
Authors:	Ravi, Shambhavi B.¹ shambhavibr.ise@bmsce.ac.in, Kurup, Sindhu¹ ksindhu.ise@bmsce.ac.in, Mohan, Banu Priya² banupriya.m@chanakyauniversity.edu.in, Nagaraj, Kalyan³ kalyan.cds@bmsce.ac.in
Source:	IAENG International Journal of Computer Science. Oct2026, Vol. 53 Issue 10, p4134-4148. 15p.
Subjects:	Swarm intelligence, Distributed computing, Embedded computer systems, Federated learning
Abstract:	Privacy-preserving data mining (PPDM) in distributed environments faces significant challenges in simultaneously achieving high utility, strong privacy guarantees, and scalability with limited computational resources. This paper introduces AAS-PPDM, the first swarm intelligence-based PPDM framework capable of solving multiagent negotiation problems with fully distributed data. The proposed system achieves high classification accuracy while maintaining a low distributional shift (measured by KL-divergence) and minimal membership inference attack (MIA) risk, without employing deep learning or requiring global coordination. Extensive experiments on the UCI Breast Cancer dataset demonstrated that AAS-PPDM attained 95.56% test accuracy at an average privacy budget of ϵ = 1.8. Compared with static uniform differential privacy (ϵ = 1.0), the proposed approach recovers approximately 82.8% of the lost utility. The system exhibits very low MIA vulnerability, approaching the performance of random guessing. Convergence analysis confirmed stability after 120 iterations, with sublinear runtime scaling and memory consumption under 200 MB, making it highly suitable for resource-constrained edge devices. The AAS-PPDM establishes a strong foundation for reliable privacy-preserving federated edge AI systems with broad applicability in healthcare, IoT, and other privacysensitive distributed domains. [ABSTRACT FROM AUTHOR]
	Copyright of IAENG International Journal of Computer Science is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Privacy-preserving data mining (PPDM) in distributed environments faces significant challenges in simultaneously achieving high utility, strong privacy guarantees, and scalability with limited computational resources. This paper introduces AAS-PPDM, the first swarm intelligence-based PPDM framework capable of solving multiagent negotiation problems with fully distributed data. The proposed system achieves high classification accuracy while maintaining a low distributional shift (measured by KL-divergence) and minimal membership inference attack (MIA) risk, without employing deep learning or requiring global coordination. Extensive experiments on the UCI Breast Cancer dataset demonstrated that AAS-PPDM attained 95.56% test accuracy at an average privacy budget of ϵ = 1.8. Compared with static uniform differential privacy (ϵ = 1.0), the proposed approach recovers approximately 82.8% of the lost utility. The system exhibits very low MIA vulnerability, approaching the performance of random guessing. Convergence analysis confirmed stability after 120 iterations, with sublinear runtime scaling and memory consumption under 200 MB, making it highly suitable for resource-constrained edge devices. The AAS-PPDM establishes a strong foundation for reliable privacy-preserving federated edge AI systems with broad applicability in healthcare, IoT, and other privacysensitive distributed domains. [ABSTRACT FROM AUTHOR]
ISSN:	1819656X