Secret Leakage Analysis For Mobile Terminal Based On N-SVDD

With the popularity of mobile terminals,real time calls,message transmission and data transmission can be conveniently carried out in the network.Office data preservation is also gradually converted from traditional paper-based storage to electronic storage.With the increase of office efficiency,it also increases the possibility that secret agents can't divulge secrets through mobile terminals,and jeopardizing the safety and privacy of state organs,government departments and enterprises.Therefore,it is a urgent problem to study a real-time and efficient secret inspection technology and establish an effective framework for analyzing the mobile terminals leakage.Based on the support vector machine(SVDD Support Vector,describing the boundary Data Description)algorithm,proposes a more suitable security field of mobile terminal data characteristics of the N-SVDD algorithm,and this algorithm is the core of the machine learning model,and proposes a mobile terminal based on N-SVDD leakage analysis framework.Through this framework,the possibility of losing the secret through the mobile terminal can be evaluated in time,and the theft secret events can be found in time.SVDD is a kind of machine learning algorithm,which can be divided into two classes of target classes.It is suitable for the scene with few samples and uniform distribution of the sample set.In the field of secrecy,the data of mobile terminals are massive and heterogeneous.Therefore,the direct application of SVDD algorithm will bring many bottlenecks,such as long execution time and unsuitable multiple outliers.To solve this problem,this paper first introduced K-Means technology and local density clustering analysis algorithm in the traditional SVDD algorithm,we propose an improved N-SVDD algorithm,to make up for the deficiency of SVDD algorithm,and thus enhance the algorithm efficiency and accuracy;and then,construct the accurate and efficient machine learning model using N-SVDD algorithm;then.Based on the proposed algorithm and model,the design and implementation of a data acquisition,including the leakage analysis model,the leakage of the mobile terminal to determine the three basic steps of the stolen key analysis framework.Finally,the performance of the N-SVDD algorithm is tested by using the unified performance evaluation standard and the actual data in the secrecy field as a test case,and the feasibility and effectiveness of thealgorithm is verified.The experimental results show that compared with the SVDD algorithm,the N-SVDD algorithm has a significant improvement in performance indexes such as time,accuracy and so on.The N-SVDD based mobile terminal theft analysis framework can be used to analyze and determine the loss of the mobile terminal efficiently and accurately.