FIELD: calculating; counting.
SUBSTANCE: invention relates to computer engineering. Disclosed is a system for classifying objects of a computer system, which comprises a) a collection means for collecting data describing an object of a computer system (hereinafter, an object); b) a convolution generating means for generating, on the basis of the feature vector describing the state of the object collected by the object state data collection means; c) a similarity degree calculation means for calculating, on the basis of the feature vector generated by the convolution generating means, using a trained model for calculating similarity parameters representing a numerical value which characterizes the probability that the classified object can belong to a given class, and the maximum degree of difference, which is a numerical value, characterizing the probability that the classified object is guaranteed to be belonging to another given class; d) an analysis means designed to make a decision on whether an object belongs to a given class, in the case when data collected before the specified collection rule has been triggered satisfy the given criterion for determining the class formed on the basis of the degree of similarity calculated by the means of calculating degrees of similarity and the limiting degree of difference, wherein said criterion is the object classification rule according to the established relationship between the degree of similarity and the maximum degree of difference.
EFFECT: technical result consists in improvement of accuracy of classification of objects of computer system due to use of two degrees of evaluation of belonging of objects of computer system to classes.
26 cl, 12 dwg, 1 tbl
Title | Year | Author | Number |
---|---|---|---|
SYSTEM AND METHOD OF DETECTING A MALICIOUS FILE | 2018 |
|
RU2739865C2 |
SYSTEM AND METHOD OF CLASSIFICATION OF OBJECTS | 2017 |
|
RU2679785C1 |
SYSTEM AND METHOD OF DETECTION OF MALICIOUS FILES USING A TRAINED MALWARE DETECTION PATTERN | 2017 |
|
RU2654151C1 |
SYSTEM AND METHOD OF MANAGING COMPUTING RESOURCES FOR DETECTING MALICIOUS FILES | 2017 |
|
RU2659737C1 |
SYSTEM AND METHOD OF MACHINE TRAINING MODEL OF DETECTING MALICIOUS FILES | 2017 |
|
RU2673708C1 |
SYSTEM AND METHOD OF SELECTING MEANS OF DETECTING MALICIOUS FILES | 2019 |
|
RU2739830C1 |
SYSTEM AND METHOD FOR TRAINING HARMFUL CONTAINER DETECTION MODEL | 2018 |
|
RU2697955C2 |
SYSTEM AND METHOD OF DETECTING MALICIOUS FILES ACCOMPANIED WITH USING THE STATIC ANALYSIS ELEMENTS | 2017 |
|
RU2654146C1 |
SYSTEM AND METHOD FOR DETECTING MALICIOUS ACTIVITY ON A COMPUTER SYSTEM | 2018 |
|
RU2697958C1 |
SYSTEM AND METHOD OF DETECTING SOURCE OF MALICIOUS ACTIVITY ON COMPUTER SYSTEM | 2018 |
|
RU2724800C1 |
Authors
Dates
2020-06-25—Published
2018-12-28—Filed