Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{192224,
        author = {Ms. Kavya Jagtap and Ms. Vedika Thakarke and Dr. Sumedh Pundkar},
        title = {FedShield: Privacy-Preserving Phishing and Fraud Detection Using Federated Learning with Client-Side SMOTE},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {3351-3360},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192224},
        abstract = {Phishing and online fraud have become major cyber- security issue, especially in e-commerce, banking and web-based services. Machine Learning based phishing detection systems exists but they are based on centralized data collection which compromises user privacy and breaches data privacy laws. Moreover, phishing datasets are naturally imbalance, causing poor minority fraudulent instance detection. To overcome this issue, this paper presents a privacy preserving phishing detection system that combines Federated Learning along with Syn- thetic Minority Oversampling Technique (SMOTE). In proposed method, raw data are kept local to participating clients, where SMOTE is performed locally to address class imbalance before model training. The model updates are aggregated at the central server by Federated Averaging Fed Avg algorithm which prevents privacy. Experimental results on a real-world phishing website dataset show that the proposed FL-SMOTE system achieves 95.30% accuracy, which is substantially higher than conventional federated learning while strictly preserving data privacy. The results show that client side SMOTE is effective in improving minority class detection with a slight performance difference as compared to centralized system. This paper presents the scalable and privacy preserving phishing detection solution.},
        keywords = {Phishing Detection, Fraud Detection, Federated Learning, Client-Side SMOTE, Privacy-Preserving, Machine Learning},
        month = {February},
        }