Android malicious application detection method based on improved gray wolf algorithm

Abstract

The invention discloses an Android malicious application detection method based on an improved gray wolf algorithm, which mainly solves the problem that the classifier modeling time is too long when the existing Android malicious application detection method based on machine learning detects a high-dimensional unbalanced data set. , the problem that malicious samples are easily misclassified. It includes the following steps: 1) Collect samples, including benign samples and malicious samples; 2) Decompile the samples, and extract the sample features to form a sample feature set; 3) Divide the sample feature set into a training set and a test set; 4) Use The improved gray wolf algorithm performs feature selection and selects the optimal feature subset; 5) Use the optimal feature subset to train the classifier. The feature selection method in the present invention can select the optimal feature subset, greatly reduces the construction time of the classifier, and can improve the detection efficiency and detection accuracy of Android malicious application detection.

Description

technical field [0001] The invention relates to the field of software security, in particular to an Android malicious application detection method based on an improved gray wolf algorithm. Background technique [0002] With the rapid development of the mobile Internet, Android has become one of the most popular mobile operating systems today. Because of its openness, Android has attracted countless developers to develop and make their own applications on it, but its huge development space also attracts a large number of hackers. Hackers use malicious applications to infringe on the property security and privacy of Android system users. Safety. Android malicious application detection has become an important means to prevent such threats. [0003] With the development of machine learning, many scholars have begun to apply machine learning methods to Android malware detection. However, when using detection methods based on machine learning, in order to obtain as much classif...

AI Technical Summary

Problems solved by technology

However, when using machine learning-based detection methods, in order to obtain as much classification information as possible, a large number of fine-grained features are usually extracted, which makes the feature dimension very high, which brings a huge workload to the machine learning classifier and reduces the performance of the classifier. classification accuracy

In addition, in the real world, the distribution of Android malicious applications and benign applications is unbalanced, the number of benign applications is far greater than the number of malicious applications, and an unbalanced data set will bias the machine learning classifier to a large class of samples, that is, benign applications , it is easy to cause malicious samples to be misclassified

Images