IOT security analysis of BDT-SVM multi-classification algorithm

Abstract

With the continuous development of the Internet of Things, the issue of IoT data security and privacy protection has received increasing attention. Compared with traditional Internet applications, IoT applications using smart terminals as supporting technologies have more complicated and serious security problems. This paper proposes a multi-classification algorithm based on double support vector machine decision tree. For all samples, the samples with the most separability were divided into two categories according to the size of the distinguishability between classes. In these two sub-categories, the most divisible samples are separately searched and divided into two categories, so that they can not be subdivided. This paper proposes a location privacy security protection mechanism based on anonymous tree and box structure, which provides location privacy protection for services oriented to intelligent terminals. The simulation results show that the common sub-collection of the anonymous group is larger, and the time overhead of building the group is smaller. At the same time, the BDT-SVM multi-classification algorithm can improve the accuracy of the intrusion detection system and reduce the detection time.