Sensor data anonymization based on genetic algorithm clustering with L-Diversity

Abstract

The collecting of digital information by various organizations is producing significant volume of data. Processing by third-party companies is requiring data to be published. Published data in its initial form typically contains sensitive information about individuals. One of ways to preserve privacy level of data and save it useful is anonymization. The paper describes a method of anonymization based on genetic algorithm clustering. It uses k-anonymity and l-diversity as privacy models which are implemented in the method. Base operators of genetic algorithm are modified to satisfy the optimization problem conditions. The experimental study focuses on investigation method application area and defines the ways of future improvement.

Keywords

• anonymization
• privacy model
• k-anonymity
• genetic algorithm
• sensor data
• clustering