Research And Implementation Of Mobile Malware Detection Based On Optimized Fuzzy C-Means

With the development of science and technology and the improvement of people’s living standard, the smart phone has become an indispensable part of people’s daily life. But some mobile phone security problems arise while smart phones bring convenience to people’s daily life. A variety of viruses, Trojans, rogue software and other malicious programs may steal user’s privacy information, or take the user’s Internet traffic, or customize service in charge privately. And such despicable acts caused leakage of user’s privacy and losses of economic, which seriously damage the vital interests of users. Therefore, how to detect and prevent the mobile phone malware is particularly urgent and important.Drawing on the experience of computer virus detection methods, the traditional mobile malware detection methods can be divided into static detection and dynamic detection, these two methods each have their own advantages and disadvantages, isolated using any single detection method cannot obtain good detection effect. With the gradual rise of the Cloud Computing and the Big Data technology, some new algorithm based on Machine Learning and Data Mining is applied to the mobile malware detection, which opens the beginning of mobile malware detection by Cloud Computing and Big Data. These detection methods combined the advantages of the static analysis and the dynamic analysis, and achieved mass malicious software automatic analysis and detection by using highly artificial intelligent engine with the help of cloud platform of high concurrency, high reliability, easy to extend and other advantages.In this paper, some improved existing Machine Learning and Data Mining algorithms are applied to the automatic classification of massive malware detection on the basis of previous studies. Firstly, proposed an Intelligent Bat Algorithm by introducing the gravitation operator to enhance the group linkage of Bat Algorithm, which is used to optimize the Fuzzy C-Means clustering algorithm. Then proposed a set of mobile malware detection scheme on the Cloud platform, and designed a cloud detection system in accordance with the scheme, which could use the optimized Fuzzy C-Means clustering algorithm applied to automation analysis and detection engine, so as to achieve the massive malicious software automatic analysis and detection on Cloud. Finally, as a supplement to the cloud detection system and data source of sample, the Android client of mobile malware detection system has been designed and implemented. The experimental data show that the mobile malware detection system of Cloud Platform and Android client designed in this paper express better in detection effect than a variety of domestic security software under the premise of enough sample data.