FIELD: computer equipment.
SUBSTANCE: present technical solution relates to computer engineering, particularly to a method and system for detecting malicious files in a non-isolated medium. A computer-implemented method for detecting malicious files in a non-isolated medium comprises: a preparatory step, where: generating a collection of files, which contains at least one malicious executable file and at least one non-malicious executable file; analyzing at least one executable file, wherein: extracting data from binary and disassembled types of executable file, based on which parameters are created for further training classifier, wherein additionally by statistical method determining parameters characteristic of malicious files and / or vice versa is not malicious; wherein first and second flow graph is extracted; based on obtained parameters first and second feature vector are constructed; creating an ensemble of classifiers from: first trained classifier based on first feature vector, a second trained classifier based on a second feature vector; third classifier, trained based on first flow graph, fourth classifier, trained on second flow graph, wherein for each classifier a decision priority is determined in advance; working stage, at which: obtaining, at least, one executable file; trained at the preparatory stage classifier ensemble, to detect malicious executable files; analysis result is output.
EFFECT: disclosed is a method of detecting malicious files.
15 cl, 1 tbl, 5 dwg
Title | Year | Author | Number |
---|---|---|---|
METHOD AND SYSTEM FOR STATIC ANALYSIS OF EXECUTABLE FILES BASED ON PREDICTIVE MODELS | 2020 |
|
RU2759087C1 |
METHOD AND SYSTEM FOR SEARCHING FOR SIMILAR MALWARE BASED ON RESULTS OF THEIR DYNAMIC ANALYSIS | 2020 |
|
RU2738344C1 |
METHOD OF DETECTING MALICIOUS FILES USING LINK GRAPH | 2023 |
|
RU2823749C1 |
METHOD AND SYSTEM FOR DETERMINING MALICIOUS ACTIVITY BY ANALYZING THE BEHAVIOUR OF OBJECTS IN NON-INSULATED ENVIRONMENT | 2020 |
|
RU2743620C1 |
COMPUTER SYSTEM AND METHOD FOR DETECTING MALWARE USING MACHINE LEARNING | 2021 |
|
RU2802860C1 |
METHOD AND SYSTEM FOR CLUSTERING EXECUTABLE FILES | 2021 |
|
RU2778979C1 |
COMPUTING APPARATUS AND METHOD FOR IDENTIFYING COMPROMISED APPARATUSES BASED ON DNS TUNNELLING DETECTION | 2021 |
|
RU2777348C1 |
METHOD AND SYSTEM FOR DETERMINING SOFTWARE BELONGING BY ITS SOURCE CODE | 2019 |
|
RU2728498C1 |
SYSTEM AND METHOD OF SIMILAR FILES DETERMINING | 2015 |
|
RU2614561C1 |
METHOD AND SYSTEM FOR ELIMINATING VULNERABILITIES IN PROGRAM CODE | 2023 |
|
RU2821220C1 |
Authors
Dates
2020-06-03—Published
2020-02-21—Filed