Search
8 articles for “dynamic differential privacy”
-
Adaptive Trust-Aware Post-Quantum Secure Federated Learning with Dynamic Differential Privacy
Abstract: FL allows multiple users to train the machine learning model and ensure that they cannot share their raw data. Post-quantum secure aggregation and differential privacy for FL are introduced by the Beskar Framework; it assumes that all selected nodes will perform reliably. Due to hardware faults, network failures, or malicious intent, some of the assisting nodes may perform unreliably while we are developing it in the real-world. To assist the …
Published in Journal of Advanced Database Management & Systems · Vol. 13, Issue 2, 2026 · pp. 38–44 Read article
-
Parallel Privacy-Preserving Adaptive Federated Learning on GPU-Enabled Multi-Core Architectures
Abstract: The increasing deployment of parallel and distributed intelligent systems has intensified the need for privacy-preserving learning frameworks that can exploit multi-core and graphics processing unit (GPU)-based architectures without centralizing sensitive data. This work proposes a parallel adaptive federated learning (AFL) framework that integrates differential privacy and secure aggregation over heterogeneous multi-core and GPU platforms to enhance both data confidentiality and convergence efficiency. The framework dynamically adjusts client participation, learning rates, …
Published in Recent Trends in Parallel Computing · Vol. 13, Issue 1, 2026 · pp. 09–16 Read article
-
Analysis and Identification of Malicious Mobile Applications Using Machines Learning
Abstract: Over the past few years, malware attacks on the Android platform have surged, posing significant risks to users' financial security, personal information, and device integrity. In the first half of 2019 alone, approximately 25 million smartphones were infected, highlighting the severity of these threats. The model ranks manifest features based on their frequency in normal and malicious apps, identifying key components that distinguish benign from malicious applications. To improve detection …
Published in Journal of Microcontroller Engineering and Applications · Vol. 12, Issue 2, 2025 · pp. 17–24 Read article
-
Differential Privacy-Aware Data Sanitization for Multi-Level Security
Abstract: Multi-level security (MLS) models are fundamental for enforcing mandatory access control in high-security environments such as government, military, healthcare, and finance. However, traditional MLS frameworks, including the Bell-LaPadula and Biba models, often create rigid data silos, preventing efficient data utilization. Differential privacy (DP) presents a novel solution by enabling controlled information leakage while preserving confidentiality. By injecting statistical noise into query results, DP allows lower-clearance users to access sanitized versions …
Published in International Journal of Computer Science Languages · Vol. 3, Issue 1, 2025 · pp. 42–52 Read article
-
An Adaptive and Privacy-Aware Federated Learning Framework for Efficient and Secure Model Training Across Heterogeneous Datasets
Abstract: The problem of efficiency and privacy regarding heterogeneous data in modern distributed machine learning systems is a vital point that should be taken into account. The absence of IID data distribution, client heterogeneity, and privacy invasion during the aggregation model are the bane of conventional federated learning (FL) approaches to learning like FedAvg and FedProx. The paper proposes that the adaptive and privacy-aware FL framework (AFL-P) can be used to …
Published in Journal of Mobile Computing, Communications & Mobile Networks · Vol. 13, Issue 1, 2026 · pp. 16–25 Read article
-
Designing a Novel Insider Threat Model for Enhanced Cybersecurity
Abstract: Designing a novel insider threat model is a critical imperative in the realm of cybersecurity. As organizations face an ever-expanding threat landscape, insider threats, whether deliberate or inadvertent, present a formidable challenge to the safeguarding of sensitive data and critical assets. This abstract encapsulates the significance, challenges, and innovations inherent in crafting an effective insider threat model for enhanced cybersecurity. The necessity for novel insider threat models arises from the …
Published in International Journal of Information Security Engineering · Vol. 1, Issue 2, 2023 · pp. 24–27 Read article
-
Federated Learning for Energy Management in Next Generation Smart Cities
Abstract: Federated learning has emerged as a promising approach for addressing the challenges of energy management in next-generation smart cities. This decentralized approach to machine learning allows collaborative model training among distributed data sources, while safeguarding data privacy and security. In this study, we explore the application of federated learning techniques to optimize energy consumption, enhance grid stability, and promote sustainability in smart city environments. By aggregating data from diverse sources …
Published in Journal of Communication Engineering & Systems · Vol. 14, Issue 1, 2024 · pp. 19–27 Read article
-
Passive Digital Phenotyping for Longitudinal Burnout and Occupational Mental Health Surveillance: A Transformer-Based Explainable Deep Learning Approach Using Smartphone Behavioral Streams
Abstract: Occupational burnout constitutes a pervasive yet chronically under-surveilled public health threat, its insidious temporal evolution rendering episodic self-report instruments structurally inadequate for early detection. This paper introduces BurnoutSense, a passive digital phenotyping framework that continuously harvests eight heterogeneous smartphone behavioral data streams encompassing application usage ecology, communication metadata, geospatial mobility, screen interaction dynamics, inferred sleep rhythmicity, keystroke kinematics, ambient noise exposure, and battery/charging cadence to construct individualized multivariate behavioral signatures …
Published in International Journal of Behavioral Sciences · Vol. 3, Issue 2, 2026 · pp. 44–53 Read article