Asset information monitoring system and method for network security
Through the asset information monitoring system for network security, the attack traffic identification and vulnerability prediction module is used to solve the inefficient monitoring problems caused by artificial dependence in the existing technology, and the comprehensive coverage of network threats and timely discover security vulnerabilities are achieved, which improves the Security of asset information.
Patent Information
- Application Number
- CN202410978392.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-07-22
AI Technical Summary
In the prior art, asset information monitoring methods rely on manual processing, are inefficient and difficult to fully cover cyber threats, resulting in the neglect of security vulnerabilities or delayed discovery.
A network security-oriented asset information monitoring system is designed, including an attack traffic identification module, asset information security risk prediction module and asset information security risk warning module. By capturing messages from network devices, extracting protocol characteristics, identifying network device types and Attack traffic attribute types, preprocess vulnerability text data, generate vulnerability data sets, conduct vulnerability category prediction and risk assessment, and issue security risk warnings in a timely manner.
It improves the efficiency of asset information monitoring, can fully cover all network security threats, promptly detect security vulnerabilities, and improves the security of asset information.
Smart Images

Figure CN118509267B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of network security, and in particular to an asset information monitoring system and method for network security. Background Art
[0002] With the rapid development of information technology, enterprises and various organizations are increasingly dependent on network assets. Network assets, including physical assets (such as servers, switches, time clock machines, health code recognition devices, etc.) and virtual assets (such as databases, middleware, etc.), have become an indispensable infrastructure to support modern business operations. However, while these assets bring convenience, they also face severe security challenges. Security incidents such as cyber attacks, data leaks, and malicious code intrusions occur frequently, seriously threatening the security and stability of network assets.
[0003] In the existing technology, asset information monitoring methods often fail to update asset information in a timely manner, and some monitoring methods rely on manual intervention with a low degree of automation. This is not only inefficient but also prone to errors. When dealing with large-scale network assets, this manual-dependent approach is obviously difficult to meet the needs. Furthermore, network attack methods are constantly being updated, and security threats are becoming increasingly complex and changeable. Traditional monitoring methods often cannot fully cover all threats, resulting in security vulnerabilities being ignored or delayed in discovery.
[0004] Therefore, there is an urgent need for an asset information monitoring system and method for network security to solve the above problems. Summary of the invention
[0005] The purpose of the present invention is to provide an asset information monitoring system and method for network security: to solve the technical problems in existing solutions that asset information monitoring mainly relies on manual processing, resulting in low monitoring efficiency, difficulty in fully covering all threats, and security vulnerabilities being ignored or discovered late.
[0006] The purpose of the present invention can be achieved through the following technical solutions:
[0007] On the one hand, the asset information monitoring system for network security includes an attack traffic identification module, an asset information security risk prediction module, and an asset information security risk warning module;
[0008] The attack traffic identification module is used to capture the messages sent by the network device, extract the protocol features based on the messages, determine the type of the network device based on the protocol features, and identify the attribute type of the attack traffic based on the type of the network device;
[0009] The asset information security risk prediction module is used to obtain vulnerability text data corresponding to the attribute type based on the attribute type of the attack traffic, obtain a vulnerability sample word list after preprocessing the vulnerability text data, generate a vulnerability data set based on the vulnerability sample word list, and predict the vulnerability category of the vulnerability data set to obtain the category to which the vulnerability belongs;
[0010] The asset information security risk warning module is used to evaluate the asset information security risk based on the category of the vulnerability, obtain the evaluation results, and issue a security risk warning for the asset information based on the evaluation results.
[0011] Furthermore, extracting the protocol features based on the message and determining the type of the network device based on the protocol features specifically includes the following processes:
[0012] Obtain the preset data attribute names in the message data based on NLP;
[0013] Calculate the information entropy of the data column corresponding to the preset data attribute name :
[0014] , where i is the data column number corresponding to the preset data attribute name, is the probability of occurrence of the data column corresponding to the preset data attribute name, It is a logarithmic function with base 2;
[0015] The data column is normalized, one-hot encoded, and dimension reduction is performed based on the principal component analysis method to obtain the protocol features of the data column. The normalization process includes:
[0016] ,in, is the normalized data, The original data is labeled The information entropy of the data column is , The maximum and minimum information entropy values of the preset data attribute name in all samples;
[0017] After adding labels based on the protocol features of the data column, the random forest algorithm is used for identification. If the identification result is SSH protocol and RPC protocol, the type of network device is terminal device. If the identification result is TELNET protocol and TLS protocol, the type of network device is application service device. If the identification result is XML protocol, the type of network device is printing device. If the identification result is RTSP protocol, the type of network device is video device.
[0018] Furthermore, identifying the attribute type of the attack traffic based on the type of network device specifically includes the following process:
[0019] Obtaining attack logs based on attack traffic, and judging based on the attack logs whether the attacker selects a default port corresponding to the type of the network device when attacking the network device, and if so, determining the attribute type of the attack traffic based on the default port, wherein each default port corresponds to an attribute type of the attack traffic;
[0020] Or, determine whether the attack log uses a forged IP address to frequently send requests to establish a TCP connection. If so, the attribute type of the attack traffic is a dos attack.
[0021] Furthermore, the vulnerability text data is preprocessed to obtain a vulnerability sample word list, and the vulnerability dataset is generated based on the vulnerability sample word list, which specifically includes the following processes:
[0022] Count the total number of vulnerability samples N and the number of vulnerability samples of each category D;
[0023] Traverse each word in the total number of vulnerability samples N, and perform statistics on word frequency, category word frequency, and category word text number;
[0024] Calculate the average distribution frequency pt of each word in the total sample and the average distribution frequency pit of each word t in category i:
[0025] ; ;in, is the frequency of word t, and fit is the frequency of word t in category i, that is, the category frequency;
[0026] Calculate the intra-class importance I of each word:
[0027] ; is the number of vulnerability samples in which word t appears in category i, that is, the number of category word texts;
[0028] Calculate the inter-class importance E of each word:
[0029] ;
[0030] Calculate the comprehensive value C of each word for the category:
[0031] ;
[0032] Sort the characteristic words of each category in descending order according to the comprehensive value;
[0033] Select the first n words in each category to form a list of characteristic words;
[0034] Extract features from the feature word list to obtain feature vectors, and generate vulnerability datasets based on the feature vectors ,in, is the feature vector of a vulnerability sample, The category to which the vulnerability sample belongs.
[0035] Furthermore, the vulnerability category prediction is performed on the vulnerability dataset to obtain the category to which the vulnerability belongs, which specifically includes the following process:
[0036] Get vulnerability dataset ;
[0037] Take each feature word node as the parent node in turn and construct j dependent classifiers;
[0038] For each dependent sub-classifier, calculate the posterior probability of the vulnerability sample :
[0039] ;in, Indicates the vulnerability category The maximum conditional probability of the vulnerability sample X, Indicates that among the total vulnerability samples The probability of a vulnerability of the class occurring;
[0040] After obtaining the posterior probability of each dependent sub-classifier, calculate the average posterior probability :
[0041] ; j is the number of dependent classifiers;
[0042] Choose the mean of the posterior probabilities The largest vulnerability category is used as the category to which the vulnerability belongs.
[0043] Furthermore, the asset information security risk is assessed based on the vulnerability category, and the assessment results are obtained by the following process:
[0044] Scan each host node in the target network based on the category of the vulnerability and identify the number of vulnerabilities WS existing on the host;
[0045] Get the proportion of asset information leakage under the category to which the vulnerability belongs , among which, the proportion of asset information leakage It is obtained by calculating the ratio of the amount of leaked asset information to the total amount of asset information;
[0046] The ratio of the number of vulnerabilities WS and asset information leakage Substitute the asset information security risk assessment coefficient into the calculation formula to obtain the asset information security risk assessment coefficient LMS, and record the asset information security risk assessment coefficient LMS as the assessment result. The asset information security risk assessment coefficient calculation formula is as follows:
[0047] ;
[0048] in, , is the weight coefficient, which is set by the system.
[0049] On the other hand, an asset information monitoring method for network security includes:
[0050] Capture the messages sent by the network device, extract the protocol features based on the messages, determine the type of network device based on the protocol features, and identify the attribute type of the attack traffic based on the type of network device;
[0051] Based on the attribute type of the attack traffic, the vulnerability text data corresponding to the attribute type is obtained, and a vulnerability sample word list is obtained after preprocessing the vulnerability text data. A vulnerability data set is generated based on the vulnerability sample word list, and the vulnerability category is predicted for the vulnerability data set to obtain the category to which the vulnerability belongs;
[0052] The asset information security risk is assessed based on the category of the vulnerability, the assessment result is obtained, and the security risk warning of the asset information is issued based on the assessment result.
[0053] Compared with the existing solutions, the present invention achieves the following beneficial effects:
[0054] The present invention captures messages sent by network devices, extracts protocol features based on the messages, determines the type of network devices based on the protocol features, and identifies the attribute type of attack traffic based on the type of network devices; obtains vulnerability text data corresponding to the attribute type based on the attribute type of the attack traffic, obtains a vulnerability sample word list after preprocessing the vulnerability text data, generates a vulnerability data set based on the vulnerability sample word list, predicts the vulnerability category of the vulnerability data set, and obtains the category to which the vulnerability belongs; evaluates the security risk of asset information based on the category to which the vulnerability belongs, obtains an evaluation result, and issues a security risk warning for asset information based on the evaluation result.
[0055] The present invention can improve the efficiency of asset information monitoring, can comprehensively cover all network security threats, timely discover security loopholes, and improve asset information security. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0057] Figure 1is a system block diagram of a network security-oriented asset information monitoring system according to an embodiment of the present invention;
[0058] Figure 2 The present invention is a flowchart of a method for monitoring network security-oriented asset information. DETAILED DESCRIPTION
[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0060] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.
[0061] This embodiment provides an asset information monitoring system for network security. Figure 1 is a system block diagram of an asset information monitoring system for network security according to an embodiment of the present invention. Figure 1 As shown, the system includes an attack traffic identification module, an asset information security risk prediction module and an asset information security risk warning module;
[0062] The attack traffic identification module is used to capture the messages sent by the network device, extract the protocol features based on the messages, determine the type of the network device based on the protocol features, and identify the attribute type of the attack traffic based on the type of the network device;
[0063] The asset information security risk prediction module is used to obtain vulnerability text data corresponding to the attribute type based on the attribute type of the attack traffic, obtain a vulnerability sample word list after preprocessing the vulnerability text data, generate a vulnerability data set based on the vulnerability sample word list, and predict the vulnerability category of the vulnerability data set to obtain the category to which the vulnerability belongs;
[0064] The asset information security risk warning module is used to evaluate the asset information security risk based on the category of the vulnerability, obtain the evaluation results, and issue a security risk warning for the asset information based on the evaluation results.
[0065] In summary, the present invention captures the message sent by the network device, extracts the protocol features based on the message, determines the type of the network device based on the protocol features, and identifies the attribute type of the attack traffic based on the type of the network device; obtains the vulnerability text data corresponding to the attribute type based on the attribute type of the attack traffic, obtains the vulnerability sample word list after preprocessing the vulnerability text data, generates a vulnerability data set based on the vulnerability sample word list, predicts the vulnerability category of the vulnerability data set, and obtains the category to which the vulnerability belongs; evaluates the asset information security risk based on the category to which the vulnerability belongs, obtains the evaluation result, and performs a security risk warning for the asset information based on the evaluation result. The present invention can improve the efficiency of asset information monitoring, can comprehensively cover all network security threats, timely discover security vulnerabilities, and improve the security of asset information.
[0066] It is worth mentioning that you can use the tshark tool to capture packets sent by network devices. Tshark is the command-line version of Wireshark, a powerful network protocol analyzer. It allows users to capture packet data from a live network, or read packets from previously saved capture files, and can decode and output these packets in a variety of formats. The functionality of tshark is similar to tcpdump, but provides more flexibility and analysis capabilities.
[0067] In some embodiments, extracting protocol features based on the message and determining the type of the network device based on the protocol features specifically includes the following process:
[0068] Obtain the preset data attribute names in the message data based on NLP (Natural Language Processing);
[0069] Calculate the information entropy of the data column corresponding to the preset data attribute name :
[0070] , where i is the data column number corresponding to the preset data attribute name, is the probability of occurrence of the data column corresponding to the preset data attribute name, It is a logarithmic function with base 2;
[0071] The data column is normalized, one-hot encoded, and dimension reduction is performed based on the principal component analysis method to obtain the protocol features of the data column. The normalization process includes:
[0072] ,in, is the normalized data, The original data is labeled The information entropy of the data column is , The maximum and minimum information entropy values of the preset data attribute name in all samples;
[0073] Among them, one-hot encoding, also known as single-bit effective encoding or unique hot encoding, is a commonly used method to convert categorical variables into a numerical form that can be understood by machine learning algorithms, and can generate a new feature for each value of the sample.
[0074] Based on the principal component analysis method, the dimensionality reduction is performed, and the protocol characteristics of the data column are obtained, including:
[0075] Data standardization: First, the original data is standardized so that each feature has zero mean and unit variance. This step is to eliminate the impact of different dimensions on data analysis.
[0076] Calculate the covariance matrix: Then, calculate the covariance matrix of the standardized data. The covariance matrix measures the linear relationship between different features and is the basis for subsequent eigenvalue decomposition.
[0077] Solve the eigenvalues and eigenvectors of the covariance matrix: Next, perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues and corresponding eigenvectors. The size of the eigenvalue reflects the degree of data variation in the direction of the corresponding eigenvector, that is, the amount of information.
[0078] Select principal components: Sort by eigenvalues and select the eigenvectors corresponding to the first k largest eigenvalues as principal components. Here k is the dimension of the new feature space, which is usually determined based on actual needs or the proportion of explained variance.
[0079] Data transformation: Finally, the selected feature vectors are used to transform the original data into a new feature space, thus achieving data dimensionality reduction.
[0080] Get the protocol characteristics of the data column:
[0081] In the data column obtained after dimensionality reduction, each data point is the representation of the original data in the new feature space. These new features (i.e., principal components) have the following protocol characteristics:
[0082] Linear independence: The new features are linear combinations of the original features, but they are uncorrelated with each other, that is, the covariance is 0. This makes the new feature space more concise and efficient.
[0083] Information retention: The new feature space retains the main information of the original data as much as possible. Specifically, the first b principal components contain most of the information in the original data (usually measured by the proportion of explained variance).
[0084] Orthogonality: From the perspective of linear algebra, the eigenvectors (i.e., principal components) in the new feature space are orthogonal, which means that there is no linear relationship between them, thus avoiding information redundancy.
[0085] Dimensionality reduction effect: Through dimensionality reduction, the high-dimensional features of the original data are mapped to a low-dimensional space, reducing the complexity of the data and the computational cost. At the same time, the reduced-dimensional data is easier to visualize and further process.
[0086] After adding labels based on the protocol features of the data column, the random forest algorithm is used for identification. If the identification results are SSH (Secure Shell) protocol and RPC (Remote Procedure Call Protocol), the type of network device is a terminal device. If the identification results are TELNET protocol and TLS (Transport Layer Security) protocol, the type of network device is an application service device. If the identification result is XML (Extensible Markup Language) protocol, the type of network device is a printing device. If the identification result is RTSP (Real Time Streaming Protocol), the type of network device is a video device.
[0087] Identification using the random forest algorithm includes:
[0088] 1. Data Preparation
[0089] Collect data: Collect a data set containing various protocol characteristics. These characteristics may include packet size, transmission frequency, frequency of occurrence of specific byte sequences, port numbers, protocol header information, etc.
[0090] Preprocess the data: clean the data, handle missing values, and possibly normalize or standardize the data to ensure that features of different dimensions can be compared fairly in the model.
[0091] Define labels: Assign a label to each data sample, which indicates the protocol type to which the data sample belongs.
[0092] 2. Feature extraction
[0093] Feature selection: Extract features useful for protocol identification from the raw data. This may be done based on domain knowledge or feature selection algorithms (such as information gain, chi-square test, etc.).
[0094] Feature transformation: If necessary, transform the original features into a form that is more suitable for model processing. For example, convert categorical variables into One-Hot Encoding.
[0095] 3. Divide the dataset
[0096] Training set and test set: Divide the data set into training set and test set. Usually 70%-80% of the data is used as the training set, and the remaining data is used as the test set.
[0097] 4. Training the Random Forest Model
[0098] Selection of Random Forest: The decision to use the Random Forest algorithm was made because of its ability to handle high-dimensional data, robustness to outliers and noisy data, and its ability to automate feature selection.
[0099] Set parameters: Configure the parameters of the random forest, such as the number of trees (n_estimators), the maximum depth of the tree (max_depth), the minimum number of samples required to split an internal node (min_samples_split), etc.
[0100] Train the model: Use the training set data to train the random forest model.
[0101] 5. Model Evaluation
[0102] Test model: Use the test set to evaluate the performance of the model. Common evaluation indicators include accuracy, recall, F1 score, etc.
[0103] Adjust parameters: Adjust the parameters of the random forest based on the test results to improve the performance of the model.
[0104] In some embodiments, identifying the attribute type of attack traffic based on the type of network device specifically includes the following process:
[0105] Obtaining attack logs based on attack traffic, and judging based on the attack logs whether the attacker selects a default port corresponding to the type of the network device when attacking the network device, and if so, determining the attribute type of the attack traffic based on the default port, wherein each default port corresponds to an attribute type of the attack traffic;
[0106] Or, determine whether the attack log uses a forged IP address to frequently send requests to establish a TCP (Transmission Control Protocol) connection. If so, the attribute type of the attack traffic is a dos attack.
[0107] In some embodiments, the vulnerability text data is preprocessed to obtain a vulnerability sample word list, and generating a vulnerability data set based on the vulnerability sample word list specifically includes the following process:
[0108] Count the total number of vulnerability samples N and the number of vulnerability samples of each category D;
[0109] Traverse each word in the total number of vulnerability samples N, and perform statistics on word frequency, category word frequency, and category word text number;
[0110] Calculate the average distribution frequency pt of each word in the total sample and the average distribution frequency pit of each word t in category i:
[0111] ; ;in, is the frequency of word t, and fit is the frequency of word t in category i, that is, the category frequency;
[0112] Calculate the intra-class importance I of each word:
[0113] ; is the number of vulnerability samples in which word t appears in category i, that is, the number of category word texts;
[0114] Calculate the inter-class importance E of each word:
[0115] ;
[0116] Calculate the comprehensive value C of each word for the category:
[0117] ;
[0118] Sort the characteristic words of each category in descending order according to the comprehensive value;
[0119] Select the first n words in each category to form a list of characteristic words;
[0120] Extract features from the feature word list to obtain feature vectors, and generate vulnerability datasets based on the feature vectors ,in, is the feature vector of a vulnerability sample, The category to which the vulnerability sample belongs.
[0121] Furthermore, the vulnerability category prediction is performed on the vulnerability dataset to obtain the category to which the vulnerability belongs, which specifically includes the following process:
[0122] Get vulnerability dataset ;
[0123] Take each feature word node as the parent node in turn and construct j dependent classifiers;
[0124] For each dependent sub-classifier, calculate the posterior probability of the vulnerability sample :
[0125] ;in, Indicates the vulnerability category The maximum conditional probability of the vulnerability sample X, Indicates that among the total vulnerability samples The probability of a vulnerability of the class occurring;
[0126] After obtaining the posterior probability of each dependent sub-classifier, calculate the average posterior probability :
[0127] ; j is the number of dependent classifiers;
[0128] Choose the mean of the posterior probabilities The largest vulnerability category is used as the category to which the vulnerability belongs.
[0129] In some embodiments, the asset information security risk is assessed based on the category to which the vulnerability belongs, and obtaining the assessment result specifically includes the following process:
[0130] Scan each host node in the target network based on the category of the vulnerability and identify the number of vulnerabilities WS existing on the host;
[0131] Get the proportion of asset information leakage under the category to which the vulnerability belongs , among which, the proportion of asset information leakage It is obtained by calculating the ratio of the amount of leaked asset information to the total amount of asset information;
[0132] It is worth noting that the amount of leaked asset information: collects specific data about the leaked incident, including the type, quantity, time, etc. of the leaked asset information. These data may come from internal security audits, external security vulnerability reports, and public media reports.
[0133] Total asset information volume: Assess the total asset information volume stored in the organization or system. This may require taking into account multiple factors, such as database size, number of users, business transaction volume, etc. In some cases, it may be necessary to estimate the total asset information volume.
[0134] The ratio of the number of vulnerabilities WS and asset information leakage Substitute the asset information security risk assessment coefficient into the calculation formula to obtain the asset information security risk assessment coefficient LMS, and record the asset information security risk assessment coefficient LMS as the assessment result. The asset information security risk assessment coefficient calculation formula is as follows:
[0135] ;
[0136] in, , is the weight coefficient, which is set by the system. Preferably, , It can be 0.6 and 0.4 respectively.
[0137] Furthermore, the asset information security risk assessment coefficient threshold is loaded to determine whether the asset information security risk assessment coefficient LMS is greater than the threshold. If so, a warning message is generated to issue a security risk warning. If not, no warning message is generated.
[0138] In some embodiments, Figure 2 is a workflow diagram of a network security-oriented asset information monitoring method according to an embodiment of the present invention, such as Figure 2 As shown, the method comprises the following steps:
[0139] Step S201: Capture the message sent by the network device, extract the protocol features based on the message, determine the type of the network device based on the protocol features, and identify the attribute type of the attack traffic based on the type of the network device;
[0140] Step S202: based on the attribute type of the attack traffic, obtain vulnerability text data corresponding to the attribute type, pre-process the vulnerability text data to obtain a vulnerability sample word list, generate a vulnerability data set based on the vulnerability sample word list, predict the vulnerability category of the vulnerability data set, and obtain the category to which the vulnerability belongs;
[0141] Step S203: Assess the security risk of asset information based on the category to which the vulnerability belongs, obtain an assessment result, and issue a security risk warning for the asset information based on the assessment result.
[0142] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.
[0143] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0144] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0145] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0146] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0147] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. The asset information monitoring system for network security is characterized by: The system includes an attack traffic identification module, an asset information security risk prediction module, and an asset information security risk warning module; The attack traffic identification module is used to capture the message sent by the network device, extract the protocol features based on the message, determine the type of the network device based on the protocol features, and identify the attribute type of the attack traffic based on the type of the network device. The process of extracting the protocol features based on the message and determining the type of the network device based on the protocol features specifically includes the following steps: Obtain the preset data attribute names in the message data based on NLP; Calculate the information entropy of the data column corresponding to the preset data attribute name : , where i is the data column number corresponding to the preset data attribute name, is the probability of occurrence of the data column corresponding to the preset data attribute name, It is a logarithmic function with base 2; The data column is normalized, one-hot encoded, and dimension reduction is performed based on the principal component analysis method to obtain the protocol features of the data column. The normalization process includes: ,in, is the normalized data, The original data is labeled The information entropy of the data column is , The maximum and minimum information entropy values of the preset data attribute name in all samples; After adding labels based on the protocol features of the data column, the random forest algorithm is used for identification. If the identification result is SSH protocol and RPC protocol, the type of network device is terminal device. If the identification result is TELNET protocol and TLS protocol, the type of network device is application service device. If the identification result is XML protocol, the type of network device is printing device. If the identification result is RTSP protocol, the type of network device is video device. The asset information security risk prediction module is used to obtain vulnerability text data corresponding to the attribute type based on the attribute type of the attack traffic, obtain a vulnerability sample word list after preprocessing the vulnerability text data, generate a vulnerability data set based on the vulnerability sample word list, and predict the vulnerability category of the vulnerability data set to obtain the category to which the vulnerability belongs. The process of obtaining a vulnerability sample word list after preprocessing the vulnerability text data and generating a vulnerability data set based on the vulnerability sample word list specifically includes the following steps: Count the total number of vulnerability samples N and the number of vulnerability samples of each category D; Traverse each word in the total number of vulnerability samples N, and perform statistics on word frequency, category word frequency, and category word text number; Calculate the average distribution frequency pt of each word in the total sample and the average distribution frequency pit of each word t in category i: ; ;in, is the frequency of word t, and fit is the frequency of word t in category i, that is, the category frequency; Calculate the intra-class importance I of each word: ; is the number of vulnerability samples in which word t appears in category i, that is, the number of category word texts; Calculate the inter-class importance E of each word: ; Calculate the comprehensive value C of each word for the category: ; Sort the characteristic words of each category in descending order according to the comprehensive value; Select the first n words in each category to form a list of characteristic words; Extract features from the feature word list to obtain feature vectors, and generate vulnerability datasets based on the feature vectors ,in, is the feature vector of a vulnerability sample, The category to which the vulnerability sample belongs; Predicting vulnerability categories for vulnerability datasets and obtaining the categories to which the vulnerabilities belong includes the following processes: Get vulnerability dataset ; Take each feature word node as the parent node in turn and construct j dependent classifiers; For each dependent sub-classifier, calculate the posterior probability of the vulnerability sample : ;in, Indicates the vulnerability category The maximum conditional probability of the vulnerability sample X, Indicates that among the total vulnerability samples The probability of a vulnerability of the class occurring; After obtaining the posterior probability of each dependent sub-classifier, calculate the average posterior probability : ; j is the number of dependent classifiers; Choose the mean of the posterior probabilities The largest vulnerability category is used as the category to which the vulnerability belongs; The asset information security risk warning module is used to evaluate the asset information security risk based on the category of the vulnerability, obtain the evaluation results, and issue a security risk warning for the asset information based on the evaluation results.
2. The network security-oriented asset information monitoring system according to claim 1, characterized in that: Identifying the attribute type of attack traffic based on the type of network device includes the following steps: Obtaining attack logs based on attack traffic, and judging based on the attack logs whether the attacker selects a default port corresponding to the type of the network device when attacking the network device, and if so, determining the attribute type of the attack traffic based on the default port, wherein each default port corresponds to an attribute type of the attack traffic; Or, determine whether the attack log uses a forged IP address to frequently send requests to establish a TCP connection. If so, the attribute type of the attack traffic is a dos attack.
3. The network security-oriented asset information monitoring system according to claim 1, characterized in that: The asset information security risk is assessed based on the vulnerability category, and the assessment results are obtained through the following process: Scan each host node in the target network based on the category of the vulnerability and identify the number of vulnerabilities WS existing on the host; Get the proportion of asset information leakage under the category to which the vulnerability belongs , among which, the proportion of asset information leakage It is obtained by calculating the ratio of the amount of leaked asset information to the total amount of asset information; The ratio of the number of vulnerabilities WS and asset information leakage Substitute the asset information security risk assessment coefficient into the calculation formula to obtain the asset information security risk assessment coefficient LMS, and record the asset information security risk assessment coefficient LMS as the assessment result. The asset information security risk assessment coefficient calculation formula is as follows: ; in, , is the weight coefficient, which is set by the system.
4. The asset information monitoring method for network security is characterized by: The asset information monitoring system for network security applicable to any one of claims 1 to 3 comprises: Capture the messages sent by the network device, extract the protocol features based on the messages, determine the type of network device based on the protocol features, and identify the attribute type of the attack traffic based on the type of network device; Based on the attribute type of the attack traffic, the vulnerability text data corresponding to the attribute type is obtained, and a vulnerability sample word list is obtained after preprocessing the vulnerability text data. A vulnerability data set is generated based on the vulnerability sample word list, and the vulnerability category is predicted for the vulnerability data set to obtain the category to which the vulnerability belongs; The asset information security risk is assessed based on the category of the vulnerability, the assessment result is obtained, and the security risk warning of the asset information is issued based on the assessment result.
