Multivariate Microaggregation with Fixed Group Size Based on the Travelling Salesman Problem
- 13 Downloads
Due to the growing use of IoT and 5G technologies, data are collected at an unprecedented pace. These data are used to improve decision-making processes. However, they could endanger individuals privacy, which is protected by international regulations. In this article, we propose a privacy-preserving microaggregation technique, inspired by the Travelling Salesman Problem, to protect individuals privacy through k-anonymity. We recall the basics on microaggregation and the TSP and, we describe the algorithm behind our approach. Also, we report experiments with real benchmark data sets showing that our approach outperforms current methods for low cardinality values.
KeywordsMicroaggregation Travelling Salesman Problem Privacy
The authors are supported by the Government of Catalonia (GC) with grant 2017-DI-002. A. Solanas is supported by the GC with project 2017-SGR-896, and by Fundació PuntCAT with the Vinton Cerf Distinction, and by the Spanish Ministry of Science & Technology with project RTI2018-095499-B-C32.
- 1.Batista E, Solanas A (2018) Process mining in healthcare: a systematic review. In: 9th International conference on information, intelligence, systems and applications. IEEE, pp 1–6Google Scholar
- 7.Solanas A, Casino F, Batista E, Rallo R (2017) Trends and challenges in smart healthcare research: a journey from data to wisdom. In: 3rd IEEE international forum on research and technologies for society and industry. Modena, Italy, pp 1–6Google Scholar
- 8.Solanas A, Martinez A (2006) VMDAV: a multivariate microaggregation with variable group size. In: 17th COMPSTAT symposium of the IASC, Rome, pp 917–925Google Scholar