2025· International Journal of Data Engineering and Intelligent Computing· Vol 8, pp. 01-18· 0 citations
TL;DR
A privacy-aware Federated Data Engineering framework that integrates federated learning, distributed data engineering, and secure model aggregation for cross-enterprise analytics and incorporates privacy-enhancing technologies to ensure confidentiality, integrity, and transparency is presented.
Abstract
The growing adoption of data-driven decision-making has improved operational intelligence, predictive analytics, and strategic planning across enterprises. However, privacy regulations, data sovereignty requirements, and competitive concerns often restrict direct data sharing between organizations. Federated Data Engineering (FDE) addresses these challenges by enabling collaborative analytics without transferring sensitive raw data. This paper presents a privacy-aware Federated Data Engineering framework that integrates federated learning, distributed data engineering, and secure model aggregation for cross-enterprise analytics. The framework supports decentralized data preprocessing, feature engineering, and encrypted parameter sharing while complying with regulations such as GDPR and HIPAA. It incorporates privacy-enhancing technologies, including differential privacy, secure multi-party computation, homomorphic encryption, and blockchain-based auditing, to ensure confidentiality, integrity, and transparency. The proposed architecture also improves scalability, communication efficiency, fault tolerance, and interoperability across heterogeneous enterprise environments. Experimental evaluation demonstrates enhanced collaborative analytics with reduced privacy risks and communication overhead. The framework provides a practical foundation for secure, trustworthy, and privacy-preserving cross-enterprise data collaboration in healthcare, finance, manufacturing, and other data-intensive industries.
As enterprise data grows across cloud, edge, and geographically distributed environments, traditional centralized analytics face challenges related to privacy, security, scalability, and regulatory compliance. To address these issues, this study proposes an Adaptive Federated Analytics Framework (AFAF) for distributed enterprise data systems. The framework enables collaborative analytics without sharing raw data by incorporating adaptive node selection, dynamic aggregation, privacy-preserving mechanisms, and communication optimization techniques. Unlike conventional federated approaches, AFAF dynamically evaluates node reliability, computational capacity, data quality, and network conditions to improve analytical performance. The framework integrates differential privacy, secure multi-party computation (SMPC), and encrypted aggregation to ensure enterprise-grade security and compliance. Experimental evaluations in hybrid cloud enterprise environments demonstrate improvements in analytical accuracy, aggregation efficiency, communication overhead, convergence stability, scalability, and fault tolerance compared with traditional federated systems. The framework also supports real-time analytics through adaptive participation thresholds and aggregation frequencies, enabling timely insights from distributed data sources. Results indicate that AFAF provides a scalable, secure, and efficient platform for privacy-preserving enterprise intelligence, with future enhancements including AI-driven orchestration, blockchain-based trust management, and autonomous analytics optimization.
K. R., B. S. Shah· International Journal of App...· 0 citations
Experimental results demonstrate high model accuracy, reduced privacy leakage, lower communication overhead, faster convergence, enhanced scalability, and strong resilience against security attacks, making the proposed framework suitable for next-generation privacy-preserving smart applications.
Seshagiri N· International Journal of Mod...· 0 citations
The findings indicate that unifying adaptive privacy preservation with decentralized integrity auditing yields a more complete cloud-security posture than either mechanism alone, and the paper outlines the empirical validation, including full-scale testbed experiments, required before deployment.
Jayakumar D, M. Ramamoorthy· International journal of com...· 0 citations
An engineering framework for privacy-preserving federated threat intelligence sharing that integrates trust-aware robust aggregation with blockchain-based integrity anchoring is proposed that confirms the effectiveness of the proposed trust-aware aggregation mechanism in mitigating malicious updates while preserving stable convergence and reliable intrusion detection performance.
Mehdi Houichi, Faouzi Jaidi, Adel Bouhoula· Journal of King Saud Univers...· 0 citations
This paper addresses the growing challenge of implementing secure, reliable, and scalable federated intelligence on consumer Internet of medical things (IoMT) devices and healthcare enterprise systems. Current solutions generally trade off individual aspects, for example, learning accuracy, cryptographic strength, and blockchain auditability, and do not present an end‐to‐end framework that ensures privacy preservation, adversarial robustness, post‐quantum security, and enterprise governance. We address these shortcomings by introducing CeN‐CHAIN, a consumer–enterprise integrated framework that incorporates post‐quantum key distribution, homomorphic encryption with minimal computational overhead, differential privacy, blockchain‐ensured validation, and federated model lifecycle management into a single architectural model. CeN‐CHAIN enables on‐device learning at a secure level, incorporating auditing, aggregability, and model custodianship in a heterogeneous IoMT and enterprise setting. Extensive experimental assessment based on evaluated healthcare IoMT dataset shows that CeN‐CHAIN attains 96.1% global accuracy, 0.958 F1‐score, and convergence in 22 rounds, which simultaneously minimizes the attack success rate (ASR) below 5% and a privacy leakage rate (PLR) below 3%. Although the framework includes sophisticated security levels, it has an approachable overhead, and the blockchain anchoring latency is 22–43 ms, energy usage of 2.15–6.15 J/FL round, and CPU usage of less than 77% in IoMT devices. These findings substantiate the claim that CeN‐CHAIN provides a balanced trade‐off between learning performance and integrated security under the evaluated healthcare IoMT experimental configuration.
D. Dhinakaran, Chin-Shiuh Shieh· International Journal of Com...· 0 citations
The paper proposes a systematic analysis framework for investigating five structural aspects, which need behavioral-specific adjustment beyond regular federated learning approaches in the following contexts: feature engineering with data locality; communication efficiency during distributed behavioral model training; differential privacy in behavioral prediction pipelines; non-IID distribution of behaviors in cross-silo federations; and secure aggregation with Byzantine resilience.
Kali Prasad Chiruvelli· International Journal of Com...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
