ABSTRACT
In the era of digital transformation, contact centers have increasingly adopted artificial intelligence (AI) to streamline customer service operations and improve user experience. However, the use of AI in processing large volumes of conversational data raises significant concerns regarding data privacy and regulatory compliance. This paper presents a novel Privacy-Preserving AI Model for the autonomous detection and masking of sensitive user data within contact center analytics. The proposed model integrates Natural Language Processing (NLP) with a differential privacy framework to identify personally identifiable information (PII), including names, addresses, social security numbers, and financial details, from both structured and unstructured data. A deep learning architecture combining Bidirectional Long Short-Term Memory (BiLSTM) and Conditional Random Fields (CRF) is employed to enhance the model’s ability to recognize context-sensitive entities in real time. To ensure data utility and minimize information loss, a context-aware masking algorithm is introduced, replacing sensitive elements with anonymized tokens while preserving semantic integrity. The model is trained on a large annotated dataset derived from real-world contact center interactions and evaluated against benchmark datasets for entity recognition and privacy performance. Experimental results demonstrate high precision and recall in PII detection, with over 94% accuracy and minimal impact on the analytic value of the conversations. Furthermore, the model adheres to key privacy standards, including GDPR and CCPA, ensuring ethical and lawful data handling. This research addresses a critical need in the AI and customer service domains by providing a scalable, compliant, and intelligent solution for privacy management. The implementation of this model can empower organizations to leverage AI-driven insights without compromising user confidentiality, thereby fostering customer trust and regulatory alignment. Future work will explore the integration of federated learning to further enhance privacy protection and model generalization across diverse linguistic and industry contexts. This study contributes to advancing the frontier of secure AI applications in enterprise-level communication systems.
References
- [1] Abdul-Azeez, O., Nwabekee, U. S., Agu, E. E., & Ijomah. (2024). Challenges and opportunities in implementing circular economy models in FMCG. International Journal of Frontline Research in Science and Technology, 3(2), 073-091. Frontline Research Journals.
- [2] Abdul-Azeez, O., Nwabekee, U. S., Agu, E. E., & Ijomah. (2024). The role of strategic business leadership in driving product marketing success: Insights from emerging markets. International Journal of Frontline Research in Science and Technology, 3(2), 001-018. Frontline Research Journals
- [3] Abdul-Azeez, T. I., Nwabekee, U. S., Agu, E. E., & Ijomah. (2024). Sustainability in product life cycle management: A review of best practices and innovations. International Journal of Applied Research in Social Sciences, 6(9), 2018-2043. Fair Eastt Publishers.
- [4] Abdul-Azeez, T. I., Nwabekee, U. S., Agu, E. E., & Ijomah. (2024). Strategic approaches to sustainability in multinational corporations: A comprehensive review. International Journal of Frontline Research in Science and Technology, 3(2), 038-054. Frontline Research Journals.
- [5] Abdul-Azeez, T. I., Nwabekee, U. S., Agu, E. E., & Ijomah. (2024). The evolution of environmental responsibility in corporate governance: Case studies and lessons learned. International Journal of Frontline Research in Science and Technology, 5(8), 20-37. Frontline Research Journals.
- [6] Abdul-Azeez, T. I., Nwabekee, U. S., Agu, E. E., & Ijomah. (2024). Brand management and market expansion in emerging economies: A comparative analysis. International Journal of Management & Entrepreneurship Research, 6(9), 2664-3596. Fair East Publishers.
- [7] Abhulimen, A. O., & Ejike, O. G. (2024). Enhancing dealership management software with AI integration for improved customer service and future innovations. International Journal of Management & Entrepreneurship Research, 6(8), 2561-2587.
- [8] Abhulimen, A. O., & Ejike, O. G. (2024). Ethical considerations in AI use for SMEs and supply chains: Current challenges and future directions. International Journal of Applied Research in Social Sciences, 6(08), 1653–1679.
- [9] Abhulimen, A. O., & Ejike, O. G. (2024). Solving supply chain management issues with AI and Big Data analytics for future operational efficiency. Computer Science & IT Research Journal, 5(8), 1780-1805.
- Abhulimen, A. O., & Ejike, O. G. (2024). Technology integration in project and event management: Empowering women entrepreneurs. International Journal of Management & Entrepreneurship Research, 6(8), 2588-2602.
- Abiagom, C. N., & Ijomah, T. I. (2024). Enhancing customer experience through AI-driven language processing in service interactions. Open Access Research Journal of Engineering and Technology.
- Abieba, O. A., Alozie, C. E., & Ajayi, O. O. (2025). Enhancing disaster recovery and business continuity in cloud environments through infrastructure as code. Journal of Engineering Research and Reports, 27(3), 127-136.
Download all article in PDF
