A Privacy-Aware and Communication-Efficient Federated Spam Detection Framework for Multi-Cloud Environments
Abstract
Abstract Spam detection in distributed multi-cloud environments presents significant challenges related to scalability, communication efficiency, and data privacy. Traditional centralized approaches are unsuitable due to strict data-sharing constraints and high communication overhead. To address these issues, this paper proposes a novel Federated Privacy-Preserving Spam Detection framework for Multi-Cloud Platforms (FPSD-MCP), which integrates federated learning with secure multi-party computation and homomorphic encryption. Unlike conventional federated approaches that rely on standard aggregation methods, FPSD-MCP introduces a privacy-aware and communication-efficient aggregation strategy that adaptively weights client updates based on data distribution, communication cost, and privacy-induced noise. This formulation enables an optimized trade-off between model accuracy, communication overhead, and privacy preservation under heterogeneous and non-IID data settings. The proposed framework is evaluated on the Enron and SpamAssassin datasets in a simulated multi-client environment and compared with baseline models including Logistic Regression, LSTM, and BERT. Experimental results demonstrate that FPSD-MCP significantly improves performance, achieving gains of 4.5%, 3.6%, and 4.1% in precision, recall, and F1-score, respectively, on the Enron dataset, and 11.7%, 7.1%, and 9.4% on the SpamAssassin dataset. These findings indicate that FPSD-MCP provides a scalable and secure…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.