Malicious Program Detection Method Based on Permission and Sensitive API
SHENG Chao;WEI Sheng-na;Information Engineering Institute, East China University of Technology;
As the Android system open source features, the phone Based on Android system is easy to become an object of attack and causes unnecessary losses to the user. In order to solve this problem, a Naive Bayesian classification detection method Based on permission feature and sensitive API is presented by this paper. This method overcomes the precondition that the characteristic attributes are assumed to be independent of each other in the Naive Bayesian classification algorithm. By extracting the permission tag of the Android application configuration file and the sensitive API in the source code, the mixed feature set is composed,and then the information gain and the chi-square test combination algorithm are used to reduce the redundant data and get the sample set suitable for the naive Bayesian classification algorithm. Finally, the Naive Bayesian classification algorithm is used to classify. The experimental results show that this method can improve the detection rate of malware and reduce the false positive rate.
【CateGory Index】： TP309