Vulnerability event level determination method and device, equipment and storage medium

By acquiring candidate vulnerability events and scoring data, using prediction models to improve the accuracy of vulnerability event level determination, solving the problem of low accuracy of level determination in the prior art, and improving the efficiency of handling network security incidents.

CN118972093APending Publication Date: 2024-11-15CHINA TELECOM CORP LTD TECHNOLOGY INNOVATION CENTER +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410940469.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-12
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

When determining the level of vulnerability event, the prior art relies on a single data analysis, resulting in low accuracy in level determination, affecting the efficiency of handling network security incidents.

Method used

By obtaining candidate vulnerability events consistent with the attribute information of the vulnerability event to be tested and their associated target data, combining the scoring data of the event database and the general vulnerability scoring system, using the prediction model to predict the handling level of vulnerability events, expanding the information dimension and data support .

Benefits of technology

It improves the accuracy of determining the level of vulnerability event, improves the efficiency of handling vulnerability events, and enhances the overall handling ability of network security events.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118972093A_ABST
    Figure CN118972093A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a vulnerability event level determination method and device, equipment and a storage medium, and relates to the technical field of computers, in particular to the technical field of network security. The method comprises the steps of obtaining at least one candidate vulnerability event consistent with attribute information of a to-be-detected vulnerability event based on the attribute information of the to-be-detected vulnerability event; obtaining target data associated with the at least one candidate vulnerability event; wherein the target data comprises event data obtained from a pre-constructed event database and score data determined through a general vulnerability score system; and based on the at least one piece of target data, predicting a target disposal level of the to-be-detected vulnerability event. According to the method, the multiple vulnerability events related to the vulnerability events are obtained in combination with the current network condition, prediction is performed from the event data and the score data, the information dimension during vulnerability event level analysis is expanded, and then the accuracy of vulnerability event level determination is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Background Art

[0002] In the network security protection activities of major tasks, it is necessary to classify the vulnerability incidents in various types of network security incidents into levels, so that network security personnel can deal with the most urgent and important vulnerability incidents in a timely manner according to the classification results, thereby improving the efficiency of handling vulnerability incidents, and then improving the efficiency of handling network security incidents, and ensuring network security in a timely manner.

[0003] At present, when determining the level of a vulnerability event: the basic data of the vulnerability event is usually analyzed to obtain the vulnerability event level. Determining the level based on a single data results in low accuracy of the determined vulnerability event level.

[0004] In summary, how to improve the accuracy of vulnerability event level determination is a technical problem that needs to be solved at present. Summary of the invention

[0005] The embodiments of the present application provide a method, apparatus, device and storage medium for determining a vulnerability event level, so as to improve the accuracy of determining the vulnerability event level.

[0006] In a first aspect, an embodiment of the present application provides a method for determining a vulnerability event level, the method comprising:

[0007] Based on the attribute information of the vulnerability event to be tested, obtaining at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be tested;

[0008] Obtain target data associated with at least one candidate vulnerability event; wherein the target data includes: event data obtained from a pre-built event database, and scoring data determined by a common vulnerability scoring system;

[0009] Based on at least one target data, a target handling level of the vulnerability event to be detected is predicted.

[0010] In a second aspect, an embodiment of the present application provides a device for determining a vulnerability event level, the device comprising:

[0011] A first acquisition unit, configured to acquire, based on the attribute information of the vulnerability event to be detected, at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be detected;

[0012] A second acquisition unit is used to acquire target data associated with at least one candidate vulnerability event; wherein the target data includes: event data acquired from a pre-built event database, and scoring data determined by a common vulnerability scoring system;

[0013] The prediction unit is used to predict a target handling level of a vulnerability event to be tested based on at least one target data.

[0014] In a possible implementation, the event database includes: at least one of a basic information database, a network-wide asset database, and a threat intelligence database; and the second acquisition unit is specifically used to:

[0015] If the event database is a basic database, obtaining basic data associated with at least one candidate vulnerability event from the basic database; wherein the basic data includes: at least one of vulnerability type, port, and vulnerability risk level;

[0016] If the event database is a network-wide asset database, then the asset data associated with at least one candidate vulnerability event is obtained from the network-wide asset database; wherein the asset data includes: the asset type and asset quantity of all attacked assets of the candidate vulnerability event;

[0017] If the event database is a threat intelligence database, then threat intelligence data associated with at least one candidate vulnerability event is obtained from the threat intelligence database; wherein the threat intelligence data includes: at least one of vulnerability implementation principle, exploitation method, and affected object.

[0018] In a possible implementation, the prediction unit is specifically used for:

[0019] Fill in the missing fields in each target data to obtain the corresponding standard data;

[0020] Abnormal standard data is eliminated from at least one standard data, and based on the remaining standard data after the elimination process, a target handling level of the vulnerability event to be tested is predicted.

[0021] In a possible implementation, the prediction unit is specifically used for:

[0022] Perform feature extraction on the remaining standard data to obtain corresponding data features, and fuse all the obtained data features to obtain fused features;

[0023] Based on the fusion features, the target disposal level of the vulnerability event to be tested is predicted through the target level prediction model; wherein the target level prediction model is obtained by training based on historical vulnerability events with disposal level annotations.

[0024] In one possible implementation, the grade prediction model is trained in the following manner:

[0025] Based on the training sample set, the training grade prediction model is trained iteratively, and when the set conditions are met, the target grade prediction model is output; wherein the following operations are performed during one cycle of iterative training:

[0026] Selecting training samples from the training sample set, the training samples include: a disposal level label of a historical vulnerability event, and a historical fusion feature determined based on other vulnerability events that are consistent with attribute information of the historical vulnerability event;

[0027] The training samples are input into the level prediction model to be trained, and the prediction labels of historical vulnerability events are obtained by combining the multi-layer decision tree in the level prediction model to be trained with the historical fusion features;

[0028] Based on the difference between the predicted label and the treatment grade label, the parameters of the grade prediction model to be trained are adjusted.

[0029] In a possible implementation, based on the difference between the prediction label and the treatment level label, the parameters of the training level prediction model are adjusted, including:

[0030] Construct a loss function based on the difference between the predicted label and the treatment level label;

[0031] Based on the loss function and the penalty function used to control the complexity of the model, the number of decision trees in the level prediction model to be trained and the iterative decision tree step size are adjusted.

[0032] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to execute the computer program to implement the steps of the method for determining the vulnerability event level provided in the embodiment of the present application.

[0033] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method for determining the vulnerability event level provided in the embodiment of the present application are implemented.

[0034] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a computer program, and the computer program is stored in a computer-readable storage medium; when the processor of an electronic device reads the computer program from the computer-readable storage medium, the processor executes the computer program, so that the electronic device executes the steps of the method for determining the vulnerability event level provided in the embodiment of the present application.

[0035] The beneficial effects of this application are as follows:

[0036] The embodiment of the present application provides a method, device, equipment and storage medium for determining the level of a vulnerability event, which relates to the field of computer technology, and in particular to the field of network security technology; in the present application, in combination with the existing network situation, based on the attribute information of the vulnerability event to be tested, at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be tested is obtained; then for each candidate vulnerability event, its associated target data is obtained, and the target data includes: event data obtained from a pre-built event library, and scoring data determined by a general vulnerability scoring system; finally, based on at least one target data, the target handling level of the vulnerability event to be tested is predicted. It can be seen that the present application not only predicts from a single dimension of the basic data of the vulnerability event, but also obtains multiple vulnerability events related to the vulnerability event in combination with the existing network situation, and predicts from the event data and scoring data, which expands the information dimension when analyzing the vulnerability event level, enriches the data support when determining the vulnerability event level, improves the accuracy of vulnerability event level determination, and thus improves the handling efficiency of vulnerability events. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 A schematic diagram of an application scenario provided for an embodiment of the present application;

[0039] Figure 2 A schematic diagram of constructing a training sample provided in an embodiment of the present application;

[0040] Figure 3 A flow chart of a level prediction model training method provided in an embodiment of the present application;

[0041] Figure 4 A schematic diagram of a specific implementation of a grade prediction model training provided in an embodiment of the present application;

[0042] Figure 5 A flow chart of a method for determining a vulnerability event level provided in an embodiment of the present application;

[0043] Figure 6 A flowchart of a specific implementation method for determining a vulnerability event level provided in an embodiment of the present application;

[0044] Figure 7 A structural diagram of a device for determining a vulnerability event level provided in an embodiment of the present application;

[0045] Figure 8A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and beneficial effects of this application clearer, the technical solution in the embodiment of this application will be clearly and completely described below in conjunction with the drawings in the embodiment of this application. Obviously, the described embodiment is only a part of the embodiment of this application, not all the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0047] It is understandable that in the following specific implementation methods of this application, when data collection or data acquisition is involved, when the various embodiments of this application are applied to specific products or technologies, relevant permissions or consents need to be obtained, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0048] Some of the terms used in the embodiments of the present application are explained below to facilitate understanding by those skilled in the art.

[0049] The Common Vulnerability Scoring System (CVSS) is an industry open standard designed to measure the severity of vulnerability incidents and help determine the urgency and importance of the required response.

[0050] Ant colony optimization (ACO), also known as ant algorithm, is a probabilistic algorithm used to find optimized paths in a graph. It is inspired by the behavior of ants in finding paths in the process of searching for food. The ant colony algorithm is a simulated evolutionary algorithm. Preliminary studies have shown that the algorithm has many excellent properties. For parameter optimization design problems, the ant colony algorithm has the effectiveness and application value of a new simulated evolutionary optimization method. In the embodiment of the present application, the ant colony algorithm is used to optimize the training grade prediction model to obtain the target grade prediction model.

[0051] The penalty function is an additional term added to the objective function during the model optimization process to control the complexity of the model, where the objective function consists of the loss function and the penalty function. The penalty function can make the model more stable and avoid overfitting. In practical applications, the appropriate penalty function is selected according to the specific situation, and the model parameters are adjusted to achieve the best model performance.

[0052] Fitness, also known as fitness value, refers to the relative ability of an individual of a known genotype to pass its genes to its offspring gene pool under certain environmental conditions. It is a measure of the individual's chances of survival and reproduction. The greater the fitness, the higher the chances of survival and reproduction. In the embodiment of the present application, the fitness value is the weight of the sample in the previous decision tree (i.e., the parent decision tree) in the grade prediction model transferred to the next decision tree (i.e., the offspring decision tree).

[0053] The word “exemplary” is used hereinafter to mean “serving as an example, example, or illustration.” Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0054] The terms "first" and "second" in this document are used for descriptive purposes only and should not be understood as explicitly or implicitly indicating relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the embodiments of this application, unless otherwise specified, the meaning of "multiple" is two or more.

[0055] The following is a brief introduction to the design concept of the embodiment of the present application:

[0056] At present, in major network security protection activities, it is necessary to classify vulnerability incidents in various types of network security incidents into different levels, so that network security personnel can deal with the most urgent and important vulnerability incidents in a timely manner according to the classification results, thereby improving the efficiency of handling vulnerability incidents, and then improving the efficiency of handling network security incidents, and ensuring network security in a timely manner.

[0057] When determining the vulnerability event level: first split the basic data of the vulnerability event and determine the level, and assign weights to the split basic data; then simply weight the level of the basic data based on the assigned weights to obtain the vulnerability event level. Since the weight assignment in this process is subjective and the data is single, the accuracy of the determined vulnerability event level is low.

[0058] In summary, how to improve the accuracy of vulnerability event level determination is a technical problem that needs to be solved at present.

[0059] In view of this, embodiments of the present application provide a method, apparatus, device and storage medium for determining vulnerability event levels, which relate to the field of computer technology, and in particular to the field of network security technology, and improve the accuracy of determining vulnerability event levels.

[0060] In the embodiment of the present application, when determining the vulnerability event level:

[0061] Firstly, based on the attribute information of the vulnerability event to be tested and in combination with the existing network situation, at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be tested is obtained; then, for each candidate vulnerability event, its associated target data is obtained, and the target data includes: event data obtained from a pre-built event library, and scoring data determined by a general vulnerability scoring system; finally, based on at least one target data, the target disposal level of the vulnerability event to be tested is predicted.

[0062] It can be seen that the implementation method for determining the vulnerability event level provided by the present application not only makes predictions from a single dimension of the basic data of the vulnerability event to be tested, but also obtains multiple vulnerability events related to the vulnerability event in combination with the existing network situation, and makes predictions from the event data and the scoring data, thereby expanding the information dimension when analyzing the vulnerability event level, enriching the data support when determining the vulnerability event level, improving the accuracy of determining the vulnerability event level, and thereby improving the efficiency of handling vulnerability events.

[0063] The following is a brief description of the application scenarios set in this application. It should be noted that the following scenarios are only used to illustrate the embodiments of this application and are not limiting. In specific implementation, the technical solutions provided in the embodiments of this application can be flexibly applied according to actual needs.

[0064] See also Figure 1 , Figure 1 A schematic diagram of an application scenario provided in an embodiment of the present application, the application scenario includes: a terminal device 110 and a server 120;

[0065] In a possible implementation, the terminal device 110 and the server 120 may be connected to each other through a communication network. The communication network may be a wired network or a wireless network. Therefore, the terminal device 110 and the server 120 may be directly or indirectly connected through a wired or wireless communication. For example, the terminal device 110 may be indirectly connected to the server 120 through a wireless access point, or the terminal device 110 may be directly connected to the server 120 through the Internet, and this application does not limit this.

[0066] Among them, the terminal device 110 includes but is not limited to mobile phones, tablet computers, laptops, desktop computers, e-book readers, intelligent voice interaction devices, smart home appliances, vehicle-mounted terminals, intelligent robots and other devices; the terminal device 110 is installed with clients and applications that support the management of digital information scanning functions.

[0067] Server 120 is the background server corresponding to the client installed in the terminal device. It can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), as well as big data and artificial intelligence platforms.

[0068] In one possible scenario, the method for determining the vulnerability event level provided in the embodiment of the present application can be executed by the terminal device 110 alone, or by the server 120 alone, or by the terminal device 110 and the server 120 in cooperation with each other, and the specific settings can be made according to actual conditions.

[0069] It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementation of the present application is not limited in this regard.

[0070] In order to improve the efficiency of vulnerability event level determination, an implementation method for automatically determining the vulnerability event level through a level prediction model is proposed in an embodiment of the present application; and in order to ensure the accuracy of vulnerability event level determination, a level prediction model training method is also provided in an embodiment of the present application to ensure the accuracy of the vulnerability time level output by the target level prediction model obtained through training.

[0071] To further illustrate the technical solution provided by the embodiment of the present application, the following takes the execution of a server alone as an example, and describes the method for training a level prediction model and the method for determining the level of a vulnerability event provided by the implementation of the present application in combination with the accompanying drawings. In addition, although the operations of the method of the present application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.

[0072] 1. Level prediction model training process

[0073] In the embodiment of the present application, the target grade prediction model is based on the training samples in the training sample set, and the target grade prediction model is trained iteratively and output when the set conditions are met. Therefore, the training samples should be constructed first.

[0074] See also Figure 2 , Figure 2 A schematic diagram of constructing a training sample is provided for the embodiment of the present application. Figure 2 It can be seen that:

[0075] First, the attribute information of the historical vulnerability events is obtained, and the attribute information includes but is not limited to: vulnerability identification (Industrial Design, ID), vulnerability uniform resource locator (Uniform Resource Locator, URL), vulnerability name; wherein the historical vulnerability events carry a disposal level label;

[0076] Secondly, based on the attribute information of the historical vulnerability events, other historical vulnerability events that are consistent with the attribute information of the historical vulnerability events are obtained from the vulnerability event information library; illustratively, other historical vulnerability events that are consistent with the vulnerability identifier of the historical vulnerability event, other historical vulnerability events that are consistent with the vulnerability URL of the historical vulnerability event, and other historical vulnerability events that are consistent with the vulnerability name of the historical vulnerability event are obtained, and at this time, all vulnerability events associated with the historical vulnerability event are obtained;

[0077] Then, the event data of all vulnerability events associated with historical vulnerability events are obtained from the pre-built event database; exemplary, basic data associated with other historical vulnerability events are obtained from the basic database, and the basic data include but are not limited to: vulnerability type, port, vulnerability risk level, impact area, attack conditions, attack method, attacker value, difficulty of exploitation, whether to deal with and other information; asset data associated with other historical vulnerability events are obtained from the whole network asset library; asset data include: asset type and asset quantity of all attacked assets in the vulnerability event, asset type includes but is not limited to: software name, software version number, hardware brand and model, component, etc.; threat intelligence data associated with other historical vulnerability events are obtained from the threat intelligence library; threat intelligence data include but are not limited to: vulnerability implementation principle, exploitation method, affected object; and,

[0078] and, a general vulnerability scoring system, which determines scoring data associated with other historical vulnerability events according to the set scoring criteria or scoring requirements;

[0079] Then, the event data and scoring data associated with all vulnerability events associated with the acquired historical vulnerability events are input and output to the data preprocessing module for data preprocessing;

[0080] Finally, the preprocessed data is input into the feature fusion module to obtain the historical fusion features associated with historical vulnerability events, and training samples are constructed based on the historical fusion features and the disposal level labels of historical vulnerability events.

[0081] After constructing the training samples, the model is trained based on the training samples. Since the training operation is the same in each loop iteration, the model training process is explained by taking one loop iteration operation as an example.

[0082] See also Figure 3 , Figure 3 A flow chart of a level prediction model training method provided in an embodiment of the present application includes the following steps:

[0083] Step S300 , selecting training samples from a training sample set, the training samples including: handling level labels of historical vulnerability events, and historical fusion features determined based on other vulnerability events consistent with attribute information of the historical vulnerability events.

[0084] It should be noted that when selecting training samples from the training sample set, a random sampling method can be used to select the training samples.

[0085] Step S301, input the training samples into the to-be-trained level prediction model, and obtain the prediction labels of the historical vulnerability events through the multi-layer decision tree in the to-be-trained level prediction model and the historical fusion features.

[0086] In an embodiment of the present application, the grade prediction model can adopt the extreme gradient boosting (eXtreme GradientBoosting, XGBoost) regression model, which is a powerful machine learning algorithm based on gradient boosting tree, suitable for solving classification and regression problems.

[0087] Step S302: adjusting the parameters of the level prediction model to be trained based on the difference between the prediction label and the treatment level label.

[0088] In one possible implementation, when adjusting the parameters of the training grade prediction model based on the difference between the prediction label and the treatment grade label: construct a loss function based on the difference between the prediction label and the treatment grade label; and adjust the number of decision trees and the iterative decision tree step size in the training grade prediction model based on the loss function and the penalty function for controlling the complexity of the model.

[0089] Optionally, based on the loss function and the penalty function for controlling model complexity, an objective function of the grade prediction model to be trained is constructed, and based on the objective function, the number of decision trees in the grade prediction model to be trained and the iterative decision tree step size are adjusted.

[0090] Among them, the objective function is:

[0091]

[0092] in, is the loss function, where N is the number of samples, l is the sample error function, and y n is the true value of the nth sample, is the predicted value of the nth sample, I is the number of trees, that is, the number of decision trees, and i represents the i-th tree in the level prediction model to be trained. is the penalty function, and defines η and λ are penalty terms, where η is the step size of the iterative decision tree and T is the total number of leaf nodes in the tree.

[0093] Considering that the level prediction model to be trained is composed of layers of decision trees, and each layer is given the highest weight to the samples of the previous layer of decision trees during training, the objective function can also be expressed as follows in the embodiment of the present application:

[0094]

[0095]

[0096] in, is the loss function, where N is the number of samples, l is the sample error function, and y t is the prediction value of the previous tree, is the predicted value for the new tree, Ω(f t ) is a penalty function used to control the complexity of the tree, cont.tant is a constant term, and η is the step size of the iterative decision tree. It should be noted that η can also be called the learning rate. The larger the η, the faster the iteration speed, the limit of the model is reached quickly, and it may not be able to converge to the true optimum. The smaller η is, the more likely it is to find a more accurate optimal value, and more space is left for the tree to be built later, but the iteration speed will be slower.

[0097] Obviously, based on the objective function and the difference between the predicted label and the treatment grade label, the number of decision trees in the grade prediction model to be trained and the iterative decision tree step size can be updated and adjusted. And the new grade prediction model is determined according to the updated and adjusted parameters. It should be noted that each decision tree can be understood as a classifier.

[0098] In an embodiment of the present application, after adjusting the parameters of the grade prediction model to be trained based on the difference value between the prediction label and the treatment grade label, the ant colony optimization algorithm is called to find the optimal solution. Through the ant colony optimization algorithm, combined with the training sample set, the number of optimized trees and the iterative decision tree step size are determined, and the grade prediction model is constructed according to the number of optimized trees.

[0099] In a possible implementation, in order to ensure the accuracy of the rank prediction model, after the rank prediction model is constructed according to the number of optimized trees, the rank prediction model is tested based on a test sample set, and the model is fine-tuned according to the test results.

[0100] See also Figure 4 , Figure 4 A specific implementation diagram of a level prediction model training provided in an embodiment of the present application, from Figure 4It can be seen that: the training samples are input into the grade prediction model to be trained, the output result of the grade prediction model to be trained is obtained, and the objective function is constructed according to the output result, and the parameters of the grade prediction model to be trained are adjusted based on the objective function.

[0101] In this application, the ant colony optimization algorithm is used to optimize the grade prediction model for model construction, which helps to capture important correlations between features and improve the practicality and accuracy of the model.

[0102] 2. Vulnerability Incident Level Determination Process

[0103] See also Figure 5 , Figure 5 A flow chart of a method for determining a vulnerability event level provided in an embodiment of the present application includes the following steps:

[0104] Step S500: based on the attribute information of the vulnerability event to be detected, obtaining at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be detected.

[0105] In a possible implementation, the attribute information includes, but is not limited to: a vulnerability identifier (Industrial Design, ID), a vulnerability uniform resource locator (Uniform Resource Locator, URL), and a vulnerability name.

[0106] Therefore, when obtaining at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be tested, based on the attribute information, obtain from the vulnerability event information library a candidate vulnerability event that is consistent with the vulnerability identifier of the vulnerability event to be tested, a candidate vulnerability event that is consistent with the vulnerability URL of the vulnerability event to be tested, and a candidate vulnerability event that is consistent with the vulnerability name of the vulnerability event to be tested, and count all the obtained candidate vulnerability events to obtain at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be tested.

[0107] Step S501, obtaining target data associated with at least one candidate vulnerability event; wherein the target data includes: event data obtained from a pre-built event database, and scoring data determined by a general vulnerability scoring system.

[0108] In a possible implementation, the pre-built event database includes but is not limited to: basic information database, network-wide asset database, and threat intelligence database. Therefore, when obtaining vulnerability events associated with each candidate vulnerability event from the pre-built event database:

[0109] Obtain basic data associated with candidate vulnerability events from the basic database; basic data includes but is not limited to: vulnerability type, port, vulnerability risk level, impact area, attack conditions, attack method, attacker value, difficulty of exploitation, whether to deal with, and other information;

[0110] Obtain asset data associated with the candidate vulnerability event from the network-wide asset library; asset data includes: asset type and asset quantity of all attacked assets in the candidate vulnerability event; asset types include but are not limited to: software name, software version number, hardware brand and model, components, etc.;

[0111] From the threat intelligence library, threat intelligence data associated with candidate vulnerability events are obtained; threat intelligence data includes but is not limited to: vulnerability implementation principle, exploitation method, and affected objects; for example, the test code that verifies the authenticity of the released vulnerability can be verified by probabilistic proof of concept (POC), i.e., the vulnerability implementation principle; the exploitation method of the vulnerability can be realized by exploit (EXP); the Common Platform Enumeration (CPE) can be used to describe which products and versions of components are affected by the vulnerability, i.e., the affected objects.

[0112] In a possible implementation, when the scoring data is determined by a common vulnerability scoring system, the common vulnerability scoring system determines the scoring data associated with the candidate vulnerability event according to a set scoring standard or scoring requirement.

[0113] In this application, the basic data, asset data, threat intelligence data of candidate events, and scoring data determined by the common vulnerability scoring system are obtained through a pre-built event database, which expands the information dimension for determining the level of vulnerability events, enriches the data information used in determination, and improves accuracy.

[0114] Step S502: predicting a target handling level of a vulnerability event to be detected based on at least one target data.

[0115] In one possible implementation, when predicting a target handling level of a vulnerability event to be tested based on at least one target data, the target data is first preprocessed, and then feature extraction is performed on the preprocessed target data to obtain fused features. Finally, the fused features are input into the trained target level prediction model to predict the target handling level of the vulnerability event to be tested.

[0116] When preprocessing the target data, firstly, the missing fields in the target data are filled with data using a random forest method to obtain corresponding standard data; then, the abnormal standard data is removed from at least one standard data, wherein the abnormal standard data is: standard information containing useless information, standard data containing abnormal values.

[0117] When extracting features from the target data after data preprocessing to obtain fused features, the ALBERT network is first used to extract features from the target data after data preprocessing to obtain corresponding data features; then the obtained data features are input into the Feature Pyramid Networks (FPN), and the data features are fused through the FPN network to obtain fused features.

[0118] Among them, the ALBERT network is an improved pre-trained language model that aims to reduce the number of parameters and accelerate training while maintaining or improving model performance; the FPN network mainly solves the multi-scale problem in object detection. Without basically increasing the computational complexity of the original model, it greatly improves the detection performance through simple changes in network connections.

[0119] In this application, the acquired target data is output to the ALBERT network for data feature extraction, and the data features are fused through the FPN network. Finally, the target level prediction model is used to predict the vulnerability disposal level to achieve automatic prediction and improve prediction speed and accuracy.

[0120] See also Figure 6 , Figure 6 A flowchart of a specific implementation method for determining a vulnerability event level provided in an embodiment of the present application includes the following steps:

[0121] Step S600, determining the attribute information of the vulnerability event to be tested;

[0122] Step S601, obtaining at least one candidate vulnerability event having attribute information consistent with the vulnerability event to be tested from a vulnerability event information library;

[0123] Step S602, obtaining basic data associated with the candidate vulnerability event from the basic database, obtaining asset data associated with the candidate vulnerability event from the whole network asset library; and obtaining threat intelligence data associated with the candidate vulnerability event from the threat intelligence library;

[0124] Step S603, the general vulnerability scoring system determines the scoring data associated with the candidate vulnerability event according to the set scoring criteria or scoring requirements;

[0125] Step S604, determining target data associated with the candidate vulnerability event based on the basic data, asset data, threat intelligence data, and scoring data;

[0126] Step S605, for missing fields in the target data, random forest method is used to fill in data to obtain corresponding standard data;

[0127] Step S606, removing the standard information containing useless information and the standard data containing abnormal values;

[0128] Step S607, using the ALBERT network to extract features from the target data remaining after the elimination process to obtain corresponding data features;

[0129] Step S608, inputting the obtained data features into the FPN network, fusing the data features through the FPN network to obtain fused features;

[0130] Step S609, based on the fusion features, the target handling level of the vulnerability event to be tested is predicted through a target level prediction model; wherein the target level prediction model is obtained by training based on historical vulnerability events with handling level annotations.

[0131] It should be noted that the execution of step S602 and step S603 is not in any particular order.

[0132] In this application, when determining the vulnerability event level, predictions are not only made from a single dimension of the basic data of the vulnerability event, but also in combination with the existing network situation, multiple vulnerability events related to the vulnerability event are obtained, and predictions are made from the event data and scoring data. This expands the information dimension when analyzing the vulnerability event level, enriches the data support when determining the vulnerability event level, improves the accuracy of vulnerability event level determination, and thereby improves the efficiency of handling vulnerability events.

[0133] Based on the same inventive concept, the present application embodiment provides a device for determining the level of a vulnerability event, see Figure 7 , the vulnerability event level determination device 700 includes:

[0134] The first acquisition unit 701 is used to acquire at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be detected based on the attribute information of the vulnerability event to be detected;

[0135] A second acquisition unit 702 is used to acquire target data associated with at least one candidate vulnerability event; wherein the target data includes: event data acquired from a pre-built event database, and scoring data determined by a universal vulnerability scoring system;

[0136] The prediction unit 703 is used to predict the target handling level of the vulnerability event to be tested based on at least one target data.

[0137] In a possible implementation, the event database includes: at least one of a basic information database, a network-wide asset database, and a threat intelligence database; the second acquisition unit 702 is specifically used to:

[0138] If the event database is a basic database, obtaining basic data associated with at least one candidate vulnerability event from the basic database; wherein the basic data includes: at least one of vulnerability type, port, and vulnerability risk level;

[0139] If the event database is a network-wide asset database, then the asset data associated with at least one candidate vulnerability event is obtained from the network-wide asset database; wherein the asset data includes: the asset type and asset quantity of all attacked assets of the candidate vulnerability event;

[0140] If the event database is a threat intelligence database, then threat intelligence data associated with at least one candidate vulnerability event is obtained from the threat intelligence database; wherein the threat intelligence data includes: at least one of vulnerability implementation principle, exploitation method, and affected object.

[0141] In a possible implementation, the prediction unit 703 is specifically configured to:

[0142] Fill in the missing fields in each target data to obtain the corresponding standard data;

[0143] Abnormal standard data is eliminated from at least one standard data, and based on the remaining standard data after the elimination process, a target handling level of the vulnerability event to be tested is predicted.

[0144] In a possible implementation, the prediction unit 703 is specifically configured to:

[0145] Perform feature extraction on the remaining standard data to obtain corresponding data features, and fuse all the obtained data features to obtain fused features;

[0146] Based on the fusion features, the target disposal level of the vulnerability event to be tested is predicted through the target level prediction model; wherein the target level prediction model is obtained by training based on historical vulnerability events with disposal level annotations.

[0147] In one possible implementation, the grade prediction model is trained in the following manner:

[0148] Based on the training sample set, the training grade prediction model is trained iteratively, and when the set conditions are met, the target grade prediction model is output; wherein the following operations are performed during one cycle of iterative training:

[0149] Selecting training samples from the training sample set, the training samples include: a disposal level label of a historical vulnerability event, and a historical fusion feature determined based on other vulnerability events that are consistent with attribute information of the historical vulnerability event;

[0150] The training samples are input into the level prediction model to be trained, and the prediction labels of historical vulnerability events are obtained by combining the multi-layer decision tree in the level prediction model to be trained with the historical fusion features;

[0151] Based on the difference between the predicted label and the treatment grade label, the parameters of the grade prediction model to be trained are adjusted.

[0152] In a possible implementation, based on the difference between the prediction label and the treatment level label, the parameters of the training level prediction model are adjusted, including:

[0153] Construct a loss function based on the difference between the predicted label and the treatment level label;

[0154] Based on the loss function and the penalty function used to control the complexity of the model, the number of decision trees in the level prediction model to be trained and the iterative decision tree step size are adjusted.

[0155] It should be noted that although several units (or modules) of the device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more units (or modules) described above can be concretized in one unit (or module). Conversely, the features and functions of one unit (or module) described above can be further divided into multiple units (or modules) for concretization. Of course, when implementing the present application, the functions of each unit (or module) can also be implemented in the same one or more software or hardware.

[0156] In the embodiments of the present application, the term unit (or module) refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the function of the module or unit.

[0157] After introducing the method and device for determining the vulnerability event level according to an exemplary embodiment of the present application, another exemplary embodiment of the present application, an electronic device, is introduced next.

[0158] Those skilled in the art will appreciate that various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to as "circuit", "module" or "system" herein.

[0159] In a possible implementation, the electronic device provided in the embodiment of the present application may include at least a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, the processor executes any step in the method for determining the vulnerability event level of various exemplary embodiments of the present application.

[0160] In this embodiment, the structure of the electronic device can be as follows Figure 8 As shown, it includes a memory 801 , a communication module 803 and one or more processors 802 .

[0161] The memory 801 is used to store computer programs executed by the processor 802. The memory 801 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and programs required for running the instant messaging function, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.

[0162] The memory 801 may be a volatile memory, such as a random-access memory (RAM); the memory 801 may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD); or the memory 801 may be any other medium that can be used to carry or store a desired computer program in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 801 may be a combination of the above memories.

[0163] The processor 802 may include one or more central processing units (CPU) or a digital processing unit, etc. The processor 802 is used to implement the above-mentioned vulnerability event level determination method when calling the computer program stored in the memory 801 .

[0164] The communication module 803 is used to communicate with the target device and other servers.

[0165] The specific connection medium between the memory 801, the communication module 803 and the processor 802 is not limited in the embodiment of the present application. Figure 8 In the embodiment, the memory 801 and the processor 802 are connected via a bus 804. The bus 804 is Figure 8The connections between the other components are described with bold lines, which are only for illustrative purposes and are not intended to be limiting. The bus 804 can be divided into an address bus, a data bus, a control bus, etc. For ease of description, Figure 8 The diagram shows that only one thick line is used, but this does not mean that there is only one bus or one type of bus.

[0166] The memory 801 stores a computer storage medium, and the computer storage medium stores computer executable instructions, and the computer executable instructions are used to implement the method for determining the vulnerability event level of the embodiment of the present application. The processor 802 is used to execute the above-mentioned method for determining the vulnerability event level.

[0167] In some possible implementations, various aspects of the method for determining vulnerability event levels provided in the present application may also be implemented in the form of a program product, which includes a computer program. When the program product is run on an electronic device, the computer program is used to enable the electronic device to execute the steps of the method for determining vulnerability event levels according to various exemplary implementations of the present application described above in this specification.

[0168] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0169] The program product of the embodiment of the present application can adopt a portable compact disk read-only memory (CD-ROM) and include a computer program, and can be run on a computing device. However, the program product of the present application is not limited to this. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with a command execution system, device or device.

[0170] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, wherein a readable computer program is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with a command execution system, apparatus, or device.

[0171] The computer program embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0172] Computer programs for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The computer program may be executed entirely on the user's computing device, partially on a terminal device, as a separate software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device via any type of network including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0173] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain a computer-usable computer program.

[0174] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program commands. These computer program commands can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable device to generate a machine, so that the commands executed by the processor of the computer or other programmable device generate commands for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that specifies functions in one or more boxes.

[0175] These computer program commands may also be stored in a computer-readable memory capable of directing a computer or other programmable device to operate in a specific manner, so that the commands stored in the computer-readable memory produce an article of manufacture comprising a command device, the command device implementing the process Figure 1 A process or multiple processes and / or boxes Figure 1 Specify the function in one or more boxes.

[0176] These computer program commands can also be loaded onto a computer or other programmable device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the commands executed on the computer or other programmable device provide the instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0177] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0178] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for determining a vulnerability event level, characterized in that: The method comprises: Based on the attribute information of the vulnerability event to be detected, obtaining at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be detected; Obtain target data associated with each of the at least one candidate vulnerability events; wherein the target data includes: event data obtained from a pre-built event database, and scoring data determined by a common vulnerability scoring system; Based on at least one of the target data, a target handling level of the vulnerability event to be detected is predicted.

2. The method according to claim 1, characterized in that The event database includes: at least one of a basic information database, a network-wide asset database, and a threat intelligence database; the acquiring of event data from a pre-built event database includes: If the event database is a basic database, obtaining basic data associated with each of the at least one candidate vulnerability events from the basic database; wherein the basic data includes: at least one of vulnerability type, port, and vulnerability risk level; If the event database is a network-wide asset database, then the asset data associated with each of the at least one candidate vulnerability event is obtained from the network-wide asset database; wherein the asset data includes: the asset type and asset quantity of all attacked assets of the candidate vulnerability event; If the event database is a threat intelligence library, the threat intelligence data associated with each of the at least one candidate vulnerability events is obtained from the threat intelligence library; wherein the threat intelligence data includes: at least one of the vulnerability implementation principle, exploitation method, and affected object.

3. The method according to claim 1, characterized in that The predicting, based on at least one of the target data, a target handling level of the vulnerability event to be detected comprises: Filling the missing fields in each of the target data with data to obtain corresponding standard data; The abnormal standard data is removed from the at least one standard data, and based on the standard data remaining after the removal process, a target handling level of the vulnerability event to be detected is predicted.

4. The method according to claim 3, characterized in that The method of predicting the target handling level of the vulnerability event to be tested based on the standard data remaining after the elimination process includes: Perform feature extraction on the remaining standard data to obtain corresponding data features, and fuse all the obtained data features to obtain fused features; Based on the fusion features, the target handling level of the vulnerability event to be tested is predicted through a target level prediction model; wherein the target level prediction model is obtained by training based on historical vulnerability events with handling level annotations.

5. The method according to claim 4, characterized in that The grade prediction model is obtained by training in the following way: Based on the training sample set, the training grade prediction model is subjected to cyclic iterative training, and when the set conditions are met, the target grade prediction model is output; wherein the following operations are performed during one cyclic iterative training process: Selecting training samples from the training sample set, the training samples including: a disposal level label of a historical vulnerability event, and a historical fusion feature determined based on other vulnerability events that are consistent with attribute information of the historical vulnerability event; Inputting the training sample into the to-be-trained grade prediction model, and obtaining the prediction label of the historical vulnerability event through the multi-layer decision tree in the to-be-trained grade prediction model and combining the historical fusion features; Based on the difference value between the prediction label and the treatment grade label, parameters of the grade prediction model to be trained are adjusted.

6. The method according to claim 5, characterized in that The step of adjusting parameters of the training grade prediction model based on the difference between the prediction label and the treatment grade label comprises: constructing a loss function based on the difference value between the predicted label and the treatment level label; Based on the loss function and the penalty function for controlling the complexity of the model, the number of decision trees and the iterative decision tree step size in the grade prediction model to be trained are adjusted.

7. A device for determining vulnerability event level, characterized in that: The device comprises: A first acquisition unit, configured to acquire, based on the attribute information of the vulnerability event to be detected, at least one candidate vulnerability event that is consistent with the attribute information of the vulnerability event to be detected; A second acquisition unit is used to acquire target data associated with each of the at least one candidate vulnerability events; wherein the target data includes: event data acquired from a pre-built event database, and scoring data determined by a universal vulnerability scoring system; The prediction unit is used to predict the target handling level of the vulnerability event to be detected based on at least one of the target data.

8. An electronic device, characterized in that: The electronic device comprises: a processor and a memory, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 6 when being executed by a processor.

Citation Information

Cited By

  • Vulnerability processing method, electronic equipment and storage medium

    CN120979793A