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{191982,
        author = {Ganesh Aappasaheb Lagad and Rohit Rajaram Kapare and Dnyaneshwar Tukaram Buddhivant and Yash Babasaheb Nirmal},
        title = {A Secure Persona Prediction Framework with Real-Time Data Leakage Prevention Using Privacy-Preserving Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {8},
        pages = {8838-8841},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=191982},
        abstract = {The increasing use of machine learning–based persona prediction systems has raised serious concerns regarding privacy leakage and regulatory compliance when processing data containing Personally Identifiable Information (PII). Conventional persona prediction models often expose sensitive attributes during training or inference, while traditional Data Leakage Prevention (DLP) mechanisms operate independently and fail to protect machine learning pipelines in real time. This paper proposes a Secure Persona Prediction System with Real-time Data Leakage Prevention, designed using privacy-by-design principles.
The proposed framework integrates synthetic data generation, anonymization, k-anonymity, and differential privacy into a unified workflow. A Random Forest classifier is employed to predict user personas, while calibrated noise injection provides protection against inference attacks in accordance with differential privacy guarantees [1]. To further mitigate leakage risks, a real-time DLP scanner continuously monitors system inputs and outputs, detecting and masking sensitive information to ensure compliance with data protection regulations such as GDPR [2] and CCPA [3]. Experimental results demonstrate that the system achieves high predictive performance while preserving privacy, preventing all detected PII leakage attempts. The proposed approach demonstrates that effective persona prediction can be achieved without compromising data privacy, making it suitable for deployment in privacy-sensitive applications [4], [5].},
        keywords = {Persona Prediction, Privacy-Preserving Machine Learning, Data Leakage Prevention (DLP), Differential Privacy, Anonymization, Personally Identifiable Information (PII), Random Forest Classifier, Synthetic Data Generation, Secure Data Analytics},
        month = {January},
        }