Research On Android Malware Detection Based On Machine Learning

With the continuous development of the mobile network,smart phones have become people's necessities in their daily life.Android-based mobile phones account for most of the market share,but Android system is becoming a platform for many malicious software breeding due to its openness.In recent years,the number of Android malware is growing at an alarming rate.Therefore,how to effectively detect Android malware has become an important issue.At present,most Android malware detection methods used by security software are signature-based detection.This method can quickly and accurately detect known malicious software but unable to detect unknown malware.In recent years,the use of machine learning methods to detect unknown Android malware has become a research hotspot.However,the detection results of this method are susceptible to factors such as data sets and detection algorithms,and the system overhead is high.Therefore,how to design the dectection scheme and how to obtain optimization of the data set become the focus of the machine-learing detection method design.This paper first introduces the background information and related research.Based on in-depth study of the Android system and its security mechanisms and comprehensive consideration of the advantages and disadvantages of traditional Android malware detection,this paper gives an malware detection scheme based on multiple features.Experimental results show that the method has a high detection rate of this paper's dataset.Furthermore,in order to adapt the current Android malware detection technology to the needs of the development of Big Data era,this paper introduces the distributed system technology,designs and implements an Android malware detection scheme based on the distributed system.Then the scheme is evaluated and analyzed through experiments.The experimental results show that the system has a high detection rate,and has a good application prospect.At last,this paper gives the summarization of the whole article and the directions for further improvement.