Android platform callback function detecting method based on machine learning method

Abstract

The invention discloses an Android platform callback function detecting method based on a machine learning method. The method includes the following steps that data sets are obtained; the obtained data sets are pre-processed; feature sets are established by using specific syntax and semantics of an Android platform, and the feature sets serves as feature vectors of the machine learning algorithm; the data sets are randomly selected to form a training set, and callback functions in the training set are manually labeled; a classifier is trained by using the training set, and a classifier model is obtained; according to the classifier model, the machine learning algorithm is adopted, all functions in the data sets are mapped to callback function sets or non-callback function sets respectively, and the identification of the callback functions is completed. The method takes the advantage of the machine learning algorithm, the callback functions and non-callback functions of the Android platform can thus be effectively and accurately identified, and the method is suitable for Android operating systems different in version.

Description

technical field [0001] The invention relates to the field of computer software, in particular to a method for detecting a callback function of an Android platform based on a machine learning method. Background technique [0002] With the rapid development of mobile application software based on the Android platform, a large number of malicious software targeting the Android operating system also follows. Malware is a kind of special software with malicious tendencies, which can endanger the security of the operating system and threaten the privacy of user data. As early as August 2010, Trojan horse viruses FakePlayer and DroidSMS based on the Android operating system began to appear, and then more malicious software became popular. These malware are not limited to stealing users' personal information, including geographical location, contact information, photos, voice, text messages, etc., but also perform various malicious operations such as making toll calls, recording ca...

AI Technical Summary

Problems solved by technology

However, regular expressions cannot be applied to different versions of the Android operating system. At the same time, the accuracy and recall are relatively low, and the universality of manually marking these regular expressions is poor.

Images