Achieving k-anonymity Using Improved Greedy Heuristics for Very Large Relational Databases

Advances in data storage, data collection and inference techniques have enabled the creation of huge databases of personal information. Dissemination of information from such databases even if formally anonymised, creates a serious threat to individual privacy through statistical disclosure. One of the key methods developed to limit statistical disclosure risk is k-anonymity. Several methods have been proposed to enforce k-anonymity notably Samarati’s algorithm and Sweeney’s Datafly, which both adhere to full domain generalisation. Such methods require a trade off between computing time and information loss. This paper describes an improved greedy heuristic for enforcing k-anonymity with full domain generalisation. The improved greedy algorithm was compared with the original methods. Metrics like information loss, computing time and level of generalisation were deployed for comparison. Results show that the improved greedy algorithm maintains a better balance between computing time and information loss.