Feature extraction method and device

Abstract

The invention provides a feature extraction method and a device, and relates to the technical field of feature extraction. The extraction method comprises the steps of conducting scanning on every source document and preset document in at least one malicious software, obtaining an original key term in every malicious software, and obtaining frequency of occurrence of each initial keyword of each original key term in every malicious software; conducting calculation of a similarity distance between every two key words in the each original key term to obtain a target distance matrix; conducting screening on each original key term according to the similarity distance recorded in the target distance matrix to obtain a target keyword, wherein the similarity distance between every two key words in the target keyword satisfies a preset category value; and finally determining that target key term is feature information of each malicious software. According to the feature extraction method and device, the technological problem in the prior art that when the malicious software is subjected to searching and killing, the searching and killing are usually not comprehensive is solved.

Description

technical field [0001] The technical field of the computer of the present invention particularly relates to a feature extraction method and device. Background technique [0002] With the advent of the Internet era, the penetration rate of smart phones in the world is also increasing. The systems of smart phones mainly include Android and IOS systems. Among them, the Android smart phone operating system has gained huge popularity by virtue of its excellent performance. market share. However, with the development of smart phones, more and more mobile malware has also appeared in the market, endangering the information security of users. Major security laboratories are gradually focusing on mobile phone security protection, but how to effectively detect and kill new malware and variants of malware has always been a difficult problem. [0003] In the prior art, when detecting and killing traditional malicious software, the malware is mainly determined through a feature code ex...

AI Technical Summary

Problems solved by technology

The feature extraction method based on program binary text can only detect and kill traditional malware, but cannot detect new malware and mutant malware

Images