Securing Personal Identity in Facial Recognition: The Shift to Machine Unlearning

Machine Unlearning, which enables the successful removal of specific data from Machine Learning models, is a crucial approach for ensuring data privacy and compliance. Most existing unlearning techniques have been evaluated on conventional benchmark datasets, which often fail to replicate the complexity of real-world settings, especially when dealing with sensitive data such as private personal information. The majority of Machine Learning research has been conducted on traditional datasets, which may not fully capture the challenges encountered in real-world scenarios, particularly concerning sensitive and private data. To address this gap, the proposed method utilizes two datasets, MUFAC and MUCAC, specifically designed to evaluate the effectiveness of Machine Unlearning in facial recognition contexts. Following the training of the model using facial data, a Machine Learning framework is developed.