41 results about "Vulnerability prediction" patented technology

The invention relates to the technical field of data security, in particular to a financial data security management system and method, and the system comprises a data monitoring module, a vulnerability prediction module, a dynamic isolation module, a behavior analysis module and a security response module. According to the invention, through deep monitoring and analysis of financial data, the capability of identifying and processing abnormal operation is improved, so that the data security is enhanced, and through real-time collection of financial transaction and account behavior data and detailed analysis of fund flow frequency and account access frequency, risks are effectively identified and potential security vulnerabilities are predicted. By dynamically analyzing the association between the transaction node and the risk event, evaluating the risk level and automatically adjusting the data storage position and the access authority according to the risk evaluation, the security of the data in the transmission and storage process is remarkably improved, in addition, by monitoring the behavior mode change of the isolated account, the abnormal activity is further identified and corrected, and the safety of the data transmission and storage process is improved. And the overall safety and stability of the financial data environment are ensured.

The invention discloses a PC bridge service state and vulnerability prediction method based on the Bayesian theory, and relates to the technical field of bridge service state and vulnerability evaluation, and the method comprises the steps: building a three-dimensional finite element model of a bridge; key parameters influencing the service state of the bridge are determined, the three-dimensional finite element model is operated, the mid-span deflection value and the average prestress loss value of each bridge sample at different moments are obtained, and a database is formed; a GRU proxy model for prediction is obtained through database training; aiming at a to-be-studied bridge, obtaining an actually-measured midspan deflection measured value of the bridge; using the key parameters of the bridge and the GRU agent model to obtain a mid-span deflection prediction value of the bridge; performing Bayesian parameter estimation according to the measured value and the predicted value of the mid-span deflection of the bridge, and estimating bridge parameters of the bridge, which conform to posterior probability distribution; and substituting the estimated bridge parameters conforming to the posterior probability distribution into the GRU agent model to obtain a time-varying curve of the mid-span deflection value and the prestress average loss value of the bridge, and a time-varying vulnerability curve of the bridge. According to the method, the relatively accurate bridge key parameters are calculated through Bayesian parameter estimation, and more accurate prediction can be achieved.

The invention provides a security vulnerability association analysis method and device, equipment and a program product, and belongs to the field of industrial internet. The method comprises the following steps: acquiring attribute information of each physical node in a plurality of physical nodes of the industrial internet; determining vulnerability combination information based on the attribute information of each physical node, a network topological graph of the industrial internet and a plurality of vulnerability prediction models; wherein each vulnerability prediction model corresponds to one vulnerability element, and the vulnerability elements comprise security vulnerabilities or defense mechanisms of the security vulnerabilities; each vulnerability prediction model is used for predicting whether the plurality of physical nodes have a corresponding vulnerability element according to the attribute information of the plurality of physical nodes; the vulnerability combination information is used for combining and representing vulnerability elements in a plurality of physical nodes; and performing association analysis based on the vulnerability combination information, and determining an attack path in the industrial internet. The method is suitable for an industrial internet security protection process, and is used for solving the problem of low vulnerability analysis prediction accuracy.

The invention is suitable for the technical field of vulnerability security management, and provides a full-life-cycle vulnerability security management method and system, and the method comprises the steps: recognizing potential security vulnerabilities in a system based on active scanning and passive monitoring; preprocessing the related data of the identified potential security vulnerabilities, inputting the preprocessed data into a trained Transform model, extracting features by using a self-attention mechanism, and predicting from a time sequence, an affected asset type and a vulnerability type to obtain a vulnerability prediction report; formulating a vulnerability repair strategy based on the vulnerability prediction report, and then performing repair according to the formulated strategy; the repaired vulnerability is verified by combining an automatic tool and manual confirmation, a comprehensive vulnerability prevention and management strategy is generated, full life cycle management from vulnerability discovery, prediction and repair to subsequent prevention is covered, and vulnerability analysis accuracy and vulnerability repair efficiency are effectively improved.

The invention relates to the technical field of software security, in particular to a decentralized financial platform loan intelligent contract vulnerability detection method and system based on a graph neural network, and the method comprises the steps: collecting intelligent contract source code data, and carrying out the vulnerability marking of the intelligent contract source code data according to whether there is a vulnerability in the data; compiling intelligent contract source code data into an abstract syntax tree, and constructing a contract code heterogeneous graph; constructing a graph neural network model, and performing vulnerability prediction training on the graph neural network model by using the contract code heterogeneous graph to obtain an intelligent contract vulnerability detection model, the graph neural network model captures structural features of a contract code heterogeneous graph layer by layer by using a hierarchical attention enhancement mechanism and predicts and identifies vulnerabilities in contract codes based on the structural features; and converting the to-be-detected smart contract source code into a target contract code heterogeneous graph to perform vulnerability prediction on the target contract code heterogeneous graph. According to the method, the semantics and data flow characteristics of the contract can be accurately captured, and efficient and accurate detection of the intelligent contract vulnerability is realized.

The embodiment of the invention provides a vulnerability risk prediction method and device, equipment and a medium, and is used for solving the problems that in the prior art, vulnerability prediction efficiency is low, and the risk level of vulnerabilities cannot be accurately determined. In the method, a to-be-detected SBOM is matched with a vulnerability database, and a target vulnerability existing in the to-be-detected SBOM and a target component associated with the target vulnerability are determined; determining a first score corresponding to the severity of the target vulnerability; determining a second score corresponding to a target dependency relationship between the item and the target component; determining a third score corresponding to the target quotation frequency of the item quoting the target component; determining a risk score of the target vulnerability according to the first score, the second score and the third score; based on the risk score of the target vulnerability, whether the target vulnerability is processed preferentially is determined, the vulnerability risk prediction efficiency is improved, and the vulnerability risk level can be determined more accurately by performing multi-dimensional assessment on the vulnerability risk.

The invention discloses a big data vulnerability monitoring and data maintenance method and system based on cloud computing. The method comprises the following steps: S1, constructing a spatio-temporal joint feature set; s2, constructing a dynamic trust graph based on the spatio-temporal joint feature set; s3, performing iterative updating on the dynamic trust graph by using an improved trust propagation algorithm; s4, predicting a multi-step path and diffusion possibility of the vulnerability diffused from the source node to other nodes by adopting a self-evolution diffusion Transform model; s5, a multi-head time sequence perception attention mechanism is introduced into the self-evolution diffusion Transform model, diffusion parameters are adjusted, and a vulnerability diffusion trend chart is generated; s6, performing reverse correction on the dynamic trust graph according to the vulnerability diffusion trend graph; s7, constructing a cloud native sandbox environment, and performing vulnerability verification; and S8, generating an intelligent repair plan in combination with the historical patch knowledge base. According to the method, an improved trust propagation algorithm and a self-evolution diffusion Transform model are fused, and a cloud environment vulnerability prediction verification closed loop is realized.

The invention discloses a large model-oriented adversarial sample iterative generation method. The method comprises the following steps of: constructing a search space based on a candidate identifier index vector; generating an adversarial example by using a Bayesian optimization framework, and dynamically selecting an optimal candidate point through a Gaussian process proxy objective function; the candidate samples are subjected to diversity sampling in combination with a determinant point process, and it is ensured that the generated confrontation examples are wide in coverage range and high in aggressiveness; and the generation process is iteratively optimized, the direction is dynamically adjusted according to attack feedback, and the generation efficiency and success rate of the confrontation examples are gradually optimized. According to the invention, by combining feedback optimization and diversity sampling technologies, the local optimum problem and the efficiency difficulty in the existing adversarial example generation method are effectively solved. Experiments show that the generated adversarial example shows a remarkable attack effect in tasks such as vulnerability prediction, clone detection and code abstracting, the robustness of the depth code model can be fully tested, and technical support is provided for model improvement and optimization.

The present invention relates to the technical field of data security, and specifically to a financial data security management system and method, including a data monitoring module, a vulnerability prediction module, a dynamic isolation module, a behavior analysis module, and a security response module. In the present invention, through in-depth monitoring and analysis of financial data, the ability to identify and handle abnormal operations is improved, thereby enhancing data security. By collecting financial transaction and account behavior data in real time, and conducting detailed analysis on the frequency of fund flow and account access, risks are effectively identified and potential security vulnerabilities are predicted. By dynamically analyzing the association between transaction nodes and risk events, the risk level is evaluated, and the data storage location and access permissions are automatically adjusted according to the risk assessment, significantly improving the security of data during transmission and storage. In addition, by monitoring the changes in the behavior patterns of accounts after isolation, abnormal activities are further identified and corrected, ensuring the overall security and stability of the financial data environment.

The invention belongs to the technical field of network security, and particularly relates to a network security penetration test platform which comprises a data acquisition layer, a data analysis layer, a vulnerability utilization layer, a report generation layer and a user interaction layer. And comprehensive penetration testing of network security in a supply chain attack scene is realized. Potential vulnerabilities of all links of the supply chain can be accurately mined, real attack evaluation vulnerability is simulated, a targeted security protection scheme is provided for enterprises, the security protection capability of the enterprise supply chain network is effectively improved, and the problems that an existing penetration test platform is insufficient in supply chain attack coverage, low in vulnerability mining efficiency and the like are solved.

The invention provides a vulnerability detection model training method, a script vulnerability detection method and a related device, and relates to the field of machine learning. The method comprises the following steps: the electronic equipment obtains vulnerability prediction loss of a sample script set according to vulnerability prediction information of the sample script set by a current first to-be-trained model; the variation degree of the sample script set is obtained, the loss is predicted according to the variation degree and the vulnerability, the current first to-be-trained model is updated, and the variation degree represents the complexity of the sample script set. And iterating the above steps until an iteration stop condition of the vulnerability detection model is satisfied. Thus, in the training process of the first to-be-trained model, the parameters of the network are automatically adjusted according to the complexity of the sample script set, so that the trained vulnerability detection model has good generalization ability and can adapt to the dynamic change and complexity of scripts.

The invention discloses a dynamic vulnerability perception-based adaptive fault tolerance method and system, and the method comprises the steps: constructing a first calculation graph based on each layer and input data of a deep learning neural network and a data flow direction, and carrying out the feature vector coding of nodes and edges of the first calculation graph; carrying out vulnerability label annotation on part of nodes after feature vector coding to obtain a training sample set, and taking unlabeled nodes as a test set; training the GNN model based on the training sample set through a loss function to obtain a vulnerability predictor model; and converting the current deep learning neural network into a second computational graph, inputting the second computational graph into the vulnerability predictor model so as to perform vulnerability node classification or marking on each node of the second computational graph, and matching corresponding fault-tolerant strategies for nodes of different vulnerability categories. According to the adaptive fault-tolerant method, the vulnerability of each layer and input data of the neural network can be dynamically sensed, and the lightweight fault-tolerant design is realized by using the vulnerability difference between the nodes.

The invention discloses an actuator vulnerability analysis method for a non-unique identifiable information physical system, and the method comprises the steps: combing an attack scene and parameters, constructing a vulnerability analysis target model set based on the information of attackers and defenders, extracting the determined line components of all possible system model parameters through matrix analysis and operation, and carrying out the vulnerability analysis. And constructing a criterion matrix of a plurality of sub-problems. And based on the property of the criterion matrix, determining the minimum number of the attack executors, and obtaining an optimal attack executor set of the sub-problem by taking whether the criterion matrix is in a full rank or not as a judgment standard through combined traversal. And generating an intersection of the optimal solutions of all the sub-problems as a global optimal solution. And constructing a simulation system to verify the defense effectiveness, and effectively attacking all states of the system. Attack and defense cognition depth alignment is achieved, and the problem that fragile point pre-judgment is not accurate is solved.

The invention relates to the technical field of communication, and particularly provides a vulnerability processing method, device and system, electronic equipment, a storage medium and a product. The method comprises the steps of obtaining network information; performing vulnerability prediction based on the network information to obtain vulnerability prediction information; processing the vulnerability prediction information by using a first model to obtain attack information; the first model is obtained by training based on a reinforcement learning principle; and attacking the target vulnerability based on the attack information. The first model automatically generates the attack information through the vulnerability prediction information, the first model is obtained by training based on the reinforcement learning principle, and the obtained attack information can be continuously adjusted according to feedback, so that the efficiency and accuracy of the attack information output by the first model are improved, and the user experience is improved. Omission or misjudgment of vulnerabilities is avoided to a certain extent, so that potential safety hazards are reduced, and the security of a network system is improved.

This application provides a vulnerability prediction method, apparatus, electronic device, and storage medium, relating to the field of network security technology, and capable of solving the problem of low security and stability in data centers. The method includes: acquiring log data from the data center, the log data carrying event content and the event level corresponding to the event content; determining vulnerability log data based on a vulnerability database and the log data, the event level corresponding to the vulnerability event content in the vulnerability log data being the vulnerability level; inputting the vulnerability event content from the vulnerability log data into a vulnerability prediction model, outputting first predicted event content, the first predicted event content having a correlation with the vulnerability event content, the event level corresponding to the first predicted event content being the vulnerability level; and the vulnerability prediction model being trained based on the vulnerability event content and the predicted event content. This application improves the efficiency and accuracy of vulnerability detection in data centers.

Aspects related to predicting zero-day vulnerabilities using anomaly detection and neural network algorithms are provided. A prediction platform may train an unsupervised algorithm for identifying suspicious packets and a prediction model for generating suspicion scores and behavior patterns based on network traffic information. The platform may segment information of packets of network traffic information into a plurality of segments. The platform may compare the segments with zero-day vulnerability information to identify known zero-day vulnerabilities. The platform may use the unsupervised algorithm to identify suspicious packets that do not correspond to known zero-day vulnerabilities. The platform may generate suspicion scores and behavior patterns for suspicious packets. The platform may further train the prediction model based on behavior patterns associated with certain suspicion scores to generate vulnerability scores. The platform may generate vulnerability scores for suspicious packets using the model. The platform may output zero-day vulnerability predictions based on the vulnerability scores.

The invention relates to a vulnerability repairing method and device and computer equipment. The method comprises the following steps: monitoring a target system; under the condition that the current version of the target system is monitored to have vulnerabilities, pulling a repair program of the target system from a software repository; and performing vulnerability repair on the target system by adopting the repair program to obtain a repair version of the target system. Wherein the software repository comprises a plurality of vulnerability repair programs, and the vulnerability repair programs are obtained according to vulnerability prediction results of historical versions of the target system by the vulnerability prediction model; the vulnerability prediction model is obtained based on a historical version repair record of the target system. By adopting the method, the system bug repairing efficiency can be improved.

The application relates to an intelligent contract vulnerability detection system based on graph convolution and global attention, which comprises a contract graph construction module, a relationship-aware graph convolution local feature extraction module, a Transformer global dependence extraction module and an intelligent contract vulnerability classification and prediction module. The contract graph construction module constructs a heterogeneous contract graph containing control flow relationships, data flow relationships and rollback interaction relationships; the relationship-aware graph convolution local feature extraction module extracts node local structure features through multi-relationship graph convolution; the Transformer global dependence extraction module captures cross-function, cross-path and long-distance implicit dependence; and the intelligent contract vulnerability classification and prediction module performs classification mapping and graph-level aggregation on the final representation of the node, and outputs contract vulnerability prediction probability and classification results. The application effectively solves the problems that mixed propagation of heterogeneous semantics is easy to cause feature interference, it is difficult to depict complex program relationships, and long-distance dependence modeling capability is insufficient.

The invention provides a network security spatial data vulnerability monitoring method based on a neural network algorithm, and relates to the technical field of network security. Inputting the to-be-tested vulnerability data into a constructed final vulnerability prediction model to obtain a vulnerability classification result and a vulnerability prediction result; and according to the vulnerability classification result and the vulnerability prediction result, monitoring the life vulnerability of the data asset life cycle in the network space. According to the invention, the problem of easy missing detection or false detection of network security monitoring in the prior art is solved.

The application discloses a bridge earthquake vulnerability prediction method and system based on machine learning, comprising: obtaining a set of earthquake parameters, constructing a nonlinear mapping model using a deep neural network, predicting vulnerability probability based on the nonlinear mapping model to obtain preliminary vulnerability probability data; obtaining bridge structure data to construct a bridge geometric model, analyzing the segmented material heterogeneity of the bridge geometric model, and if the heterogeneity exceeds a preset threshold, calculating segmented response data under seismic action; based on the segmented response data, analyzing the spatial difference vulnerability probability distribution of the bridge as a whole to obtain fine prediction results; if the deviation between the fine prediction results and the preliminary vulnerability probability exceeds a preset threshold, adjusting the nonlinear mapping model, outputting the updated vulnerability probability distribution, and obtaining the final vulnerability prediction results. The application significantly improves the accuracy and reliability of bridge earthquake vulnerability prediction, and has important significance for improving the seismic safety of bridge structures.

The present invention relates to a distribution network resilience enhancement strategy based on environmental data prediction. Comprehensive environmental data prediction and analysis are performed on the distribution network operating conditions, and the vulnerable links of the distribution network are predicted to enhance the early warning capability of disaster risks; a distribution network vulnerability prediction model based on environmental data prediction is established to improve the level of failure rate prediction; a distribution network fault carrying capacity assessment method based on environmental data is proposed; a distribution network fault carrying capacity assessment method is proposed; the change in system performance is used to reflect the impact of ice disasters on the resilience of the distribution network, and a distribution network resilience assessment model is established; the purpose is to implement planning and scheduling measures to enhance the resilience of the distribution network in the disaster defense stage and the post-disaster recovery stage, respectively, and then propose a distribution network resilience enhancement strategy under environmental data prediction. The present invention significantly improves the resilience of the distribution network, from active defense before the fault occurs to fault recovery after the disaster, and achieves an all-round improvement of the resilience of the distribution network during the disaster process and the post-disaster process through the coordination of the two means.

The invention relates to a cloud host vulnerability scanning method and device, computer equipment, a readable storage medium and a program product. The method comprises the following steps: in response to a vulnerability scanning instruction, obtaining software bill of material data of a target cloud host, and storing the software bill of material data in a graph structure; performing feature extraction on the software bill of material data to obtain multi-dimensional feature data; inputting the multi-dimensional feature data into a pre-trained graph neural network model to obtain a vulnerability prediction result, the training data of the graph neural network model including historical multi-dimensional feature data marked with a vulnerability type tag and an abnormal degree tag, and the prediction result including at least one of a vulnerability type, a vulnerability number and a vulnerability confidence; and based on the prediction result, generating and outputting a vulnerability scanning report of the target cloud host. By adopting the method, the comprehensiveness, accuracy and efficiency of vulnerability scanning can be improved.

The invention discloses an intelligent contract vulnerability detection method based on a large language model and deep learning. The method comprises the following steps: 1, obtaining and preprocessing an intelligent contract vulnerability data set; step 2, performing core code segment extraction by using ChatGPT, and performing feature extraction on the core code segment through an improved CodeBert model; 3, inputting the risk annotation data set into an intelligent contract vulnerability detection training network for training; 4, inputting the classification feature vectors into a vulnerability prediction model for vulnerability prediction; 5, evaluating the vulnerability prediction model by using the test data set; 6, performing vulnerability risk assessment according to a vulnerability prediction result; and 7, according to a vulnerability risk assessment result, providing a security repair suggestion for the smart contract and verifying a repair effect. According to the method, the big language model and the improved CodeBert model are combined, and automatic detection, evaluation and repair of the intelligent contract vulnerability are realized.

The embodiments of the present invention relate to a software vulnerability prediction method, apparatus, computer equipment and storage medium, the method comprising: obtaining at least one set of vulnerability data obtained by performing software testing on the software to be tested in the previous test cycle; inputting the at least one set of vulnerability data into a trained software vulnerability prediction model to obtain prediction parameters, the prediction parameters comprising at least a plurality of software vulnerabilities and the probability of each of the software vulnerabilities occurring in the software to be tested; determining the target software vulnerabilities that will occur in the software to be tested in the current test cycle according to the prediction parameters, by means of this method, it is possible to implement targeted and focused testing of possible vulnerabilities according to the prediction results of the model in the later software testing process, thereby improving the efficiency and effect of software testing.

The present invention provides a cross-project vulnerability detection model based on domain adaptation. It learns project code features through a deep neural network and adopts a domain adaptation method to reduce the data distribution difference between the source project and the target project. The steps include: selecting the source project according to the software code metric similarity analysis; parsing the project source code to obtain the abstract syntax tree and performing code preprocessing; building a deep learning model for model pretraining; learning the deep feature representation of the source project and the target project based on a multi-domain dataset; using the semi-supervised metric transfer learning framework (Semi-Supervised Metric Transfer Learning, SSMTL) to perform domain adaptation processing on the deep features of the source project and the target project; training a classifier based on labeled data and performing vulnerability prediction on the data in the target project. This solution is applicable to source projects with sufficient labeled vulnerability data. By utilizing the vulnerability detection model, the vulnerability detection efficiency of the target project is improved, and labor costs are reduced.

This invention provides a method, apparatus, device, medium, and program product for predicting base station vulnerabilities, relating to the field of computer technology. The method includes: preprocessing base station alarm data and performance data using base station resource data; inputting the preprocessed alarm data into a first time-series prediction model to output a first probability that the base station has vulnerabilities; inputting the preprocessed performance data into a second time-series prediction model to output a second probability that the base station has vulnerabilities; and determining the base station vulnerability information based on the first and second probabilities. This invention introduces intelligent AI model capabilities, multi-dimensionally selecting alarm data strongly correlated with base station vulnerabilities and base station performance data aggregated from a service usage perspective that directly characterizes user experience to form a vulnerability mining data system. The combination of these two systems generates a more accurate and automated base station vulnerability mining capability, thereby improving the accuracy and efficiency of base station vulnerability prediction.

The present invention discloses a dynamic vulnerability-aware adaptive fault-tolerance method and system. The method comprises: constructing a first computational graph based on the layers, input data, and data flow of a deep learning neural network, and encoding the nodes and edges of the first computational graph with feature vectors; annotating some of the feature-vector-encoded nodes with vulnerability labels to obtain a training sample set, and using the unlabeled nodes as a test set; training a GNN model based on the training sample set using a loss function to obtain a vulnerability predictor model; converting the current deep learning neural network into a second computational graph, and inputting the second computational graph into the vulnerability predictor model to classify or label each node in the second computational graph as a vulnerable node, and matching corresponding fault-tolerance strategies to nodes of different vulnerability categories. This adaptive fault-tolerance method can dynamically perceive the vulnerability of each layer of the neural network and the input data, and utilize the vulnerability differences between nodes to achieve a lightweight fault-tolerance design.

A multi-feature fusion blockchain smart contract vulnerability detection method, device, computer and storage medium based on graph neural network, which relates to the field of blockchain smart contract security detection. It solves the problem of needing a smart contract multi-version vulnerability detection method to detect whether there are vulnerabilities in mainstream smart contract versions. The method includes: obtaining a smart contract data set; processing the smart contract data set according to a smart contract vulnerability detection tool, obtaining smart contracts with vulnerabilities, and marking all smart contracts; converting all smart contract source codes into smart contract structure diagrams; extracting smart contract structure diagrams using graph neural networks, and extracting expert mode features in smart contract source codes through multi-layer perceptrons; fusing the features of the smart contract structure diagram and the expert mode features, learning the fused features through neural networks and predicting vulnerabilities; judging whether there are vulnerabilities based on the vulnerability prediction results. It is applicable to the field of smart contract vulnerability detection.

The invention discloses a code language model and code attribute graph-based source code vulnerability detection method, which comprises the following steps of: analyzing a to-be-detected source code function by using a tool to generate an initial code attribute graph, and converting a to-be-detected source code into a natural code sequence; associating the natural code sequence with the initial code attribute graph to obtain an associated code attribute graph; performing one-hot coding on the node type to obtain a node type code; inputting the associated code attribute graph into a CodeBERT model, and outputting a node content vector and a sequence representation of a function; the node type codes and the node content vectors are connected to obtain joint initial features of the nodes; inputting the joint initial features of the nodes into a GGNN model, and outputting graph-level features; and performing linear interpolation according to the sequence representation and the graph-level features of the function to obtain a vulnerability prediction result. According to the method, the code vulnerabilities can be detected more comprehensively and accurately, and the detection capability of the vulnerabilities in a complex code structure is improved.

The invention discloses an intranet equipment fingerprint processing method and system. The method comprises the following steps: acquiring service response data, equipment fingerprint data, a response data sequence, scanning data and picture information of target equipment from a pre-established communication link; analyzing and processing the data by using a large model, namely identifying service response data, and acquiring target identification information of the equipment; generating a prediction analysis list through vulnerability risk analysis of the equipment fingerprint data; performing configuration analysis on the equipment response data sequence to obtain configuration information; comparing the picture information to determine fingerprint information of the equipment; identifying an abnormal mode according to the scanning data; and inputting the identification information, the prediction analysis list, the fingerprint data and the abnormal mode into the large model, predicting an attack path, and integrating all data to generate a security assessment report. Through cooperation of a cloud server technology and a large model, the problems of complex deployment, inaccurate identification, lack of vulnerability prediction and the like are solved, and the overall security protection efficiency is improved.