An Efficient k-Anonymization Algorithm with Low Information Loss

Abstract

Publishing microdata with preserving privacy led to the paradigms of k-anonymity. Sensitive information in k-anonymous microdata cannot be linked to specific individuals with a confidence value of more than 1/k. However, k-anonymous data loses its importance with loss of precision or information contained in the data. Existing k-anonymization approaches suffer from high information loss. In this paper, we present a heuristic k-anonymization algorithm that results in very low information loss compared to existing similar algorithms. Also, the average case complexity of our algorithm is not high. Experimental results show that the information loss in our algorithm is significantly lower than that of the current state-of-the-art algorithm for k-anonymization.