Data anonymization through supervised Machine Learning

In the digital era, protecting personal data has become increasingly important as organizations collect vast amounts of sensitive information. Data anonymization techniques, such as k-anonymity, aim to strike a balance between privacy preservation and data utility. In this study, we applied supervised machine learning techniques to assess the impact of anonymization on data utility and privacy. Specifically, we evaluated k-anonymity and l-diversity models using four supervised learning methods: Naive Bayes, K-nearest neighbors, Decision Trees, and Random Forest. By measuring classification accuracy, our results highlight Naive Bayes as the most effective method for handling anonymized data, achieving high accuracy. This work provides valuable insights into the trade-offs between privacy protection and data utility, using advanced machine learning techniques to evaluate anonymization models.