Classification encryption via compressed permuted measurement matrices

Abstract

In this paper we present an efficient encryption system based on Compressive Sensing for topic detection and classification in Twitter. The proposed method first employs Joint Complexity to perform topic detection. Then based on the spatial nature of the data, we apply the theory of Compressive Sensing to perform classification from a small number of random sample measurements. The breakthrough of the method is the encryption based on the permutation of measurements which are generated when solving the classification optimization problem. The experimental evaluation with real data from Twitter presents the robustness of the encryption accuracy, without using a specific cryptographic layer, while maintaining a low computational complexity.