This paper explores the application of Privacy Preserving Data Mining (PPDM) techniques to the Adult dataset, focusing on the effects of K-anonymization. The study uses the Artificial Bee Colony (ABC) algorithm for feature generalization and suppression, aiming to remove features without compromising classification accuracy. K-anonymity is achieved through the generalization of the original dataset. The ABC algorithm optimizes the feature suppression process by identifying the best features to remove, while the generalization technique ensures that the optimal generalization range is achieved. The results show that classification accuracy, precision, and recall decrease as the k-anonymity level increases, with the proposed method maintaining a reduction in accuracy of 0 to 1.2308% as anonymity levels range from 1 to 90. The study highlights the trade-off between privacy preservation and data utility in PPDM techniques.This paper explores the application of Privacy Preserving Data Mining (PPDM) techniques to the Adult dataset, focusing on the effects of K-anonymization. The study uses the Artificial Bee Colony (ABC) algorithm for feature generalization and suppression, aiming to remove features without compromising classification accuracy. K-anonymity is achieved through the generalization of the original dataset. The ABC algorithm optimizes the feature suppression process by identifying the best features to remove, while the generalization technique ensures that the optimal generalization range is achieved. The results show that classification accuracy, precision, and recall decrease as the k-anonymity level increases, with the proposed method maintaining a reduction in accuracy of 0 to 1.2308% as anonymity levels range from 1 to 90. The study highlights the trade-off between privacy preservation and data utility in PPDM techniques.