Cross-subject data security risk assessment method and system based on deep learning

Abstract

The invention discloses a cross-subject data security risk assessment method and system based on deep learning, and the method specifically comprises the steps: collecting metadata of data interaction between different entities, and enabling the metadata and the reputation information of participating entities to generate original data; the method comprises the following steps: preprocessing original data by using an advanced feature builder, and identifying and building key feature vectors related to data security risks through a feature extraction technology; inputting the key feature vectors into a dynamic deep learning model, and continuously learning and optimizing model parameters through a back propagation algorithm; carrying out deep analysis on the key feature vector by the trained model, and outputting a security risk level and a risk probability value of each data interaction; automatically formulating or recommending data security protection measures by using a strategy intelligent recommendation system; and submitting a data interaction task through a Web interface or a mobile application program by utilizing an interactive user operation platform, checking the security risk level in real time, or obtaining the data security protection measures in real time.

Description

Technical Field

[0001] The present application relates to the field of data security and artificial intelligence technology, and mainly to a method and system for cross-subject data security risk assessment based on deep learning. Background Art

[0002] With the development of the digital economy, data plays an important role as a key production factor in various fields. Cross-subject interaction and sharing of data has become the norm, but it also brings many security risks, such as data leakage, abuse, unauthorized access, etc. Existing data security risk assessment methods mostly rely on manual analysis and traditional security rules, which are difficult to cope with complex and changing data interaction environments. Therefore, it is particularly important to develop an intelligent and automated data security risk assessment system.

[0003] In the digital age, the development of data security risk assessment technology is crucial. Existing technologies mainly include qualitative analysis based on expert experience, quantitative analysis relying on mathematical models, knowledge-based systems that simulate expert decision-making, and model-based methods that fully consider data interaction. Although these methods have certain application value in specific environments, they generally have limitations such as insufficient adaptability, limited intelligence, insufficient real-time performance, and poor user interactivity. For example, qualitative analysis methods are difficult to cope with complex and changing data environments; quantitative analysis methods require a large amount of accurate data support and high requirements for model accuracy; knowledge-based systems are limited in efficiency when processing large-scale data; and model-based methods may face the complexity of model construction and verification. Summary of the invention

[0004] In view of these limitations, the present invention proposes a cross-subject data security risk assessment system based on deep learning, which significantly improves the adaptability, intelligence level, real-time monitoring and response capabilities, and user interaction experience through an automated and intelligent risk assessment process. Using deep learning technology, the system can automatically learn from historical data and predict unknown security risks, realize real-time monitoring of the data interaction process, and quickly respond to potential security threats. In addition, the user-friendly interactive interface design enables users to easily submit data interaction tasks and obtain personalized risk assessment results and security policy recommendations. The innovation of the present invention lies in its high degree of automation, intelligence, adaptability and user interactivity, which is expected to promote the development and application of data security risk assessment technology.

[0005] According to one aspect of the present invention, a method for cross-subject data security risk assessment based on deep learning is proposed, comprising:

[0006] Capturing metadata of data interactions between different entities, collecting the data type, scale and interaction frequency of the metadata, and mining the reputation information of the participating entities, the metadata and reputation information generate raw data;

[0007] Preprocessing the raw data using an advanced feature builder, including data cleaning, normalization, and deduplication, wherein the advanced feature builder identifies and constructs key feature vectors related to data security risks through feature extraction technology;

[0008] Inputting the key feature vector into a dynamic deep learning model, wherein the dynamic deep learning model continuously learns and optimizes model parameters through a back propagation algorithm;

[0009] The trained dynamic deep learning model conducts in-depth analysis of the input key feature vectors, intelligently assesses the security risk level of each data interaction, and outputs a quantitative risk probability value;

[0010] According to the risk probability value, a strategy intelligent recommendation system is used to automatically formulate or recommend data security protection measures, wherein the strategy intelligent recommendation system includes data encryption, access permission control and data desensitization;

[0011] Utilize the interactive user operation platform to submit data interaction tasks through the web interface or mobile application, view the security risk level in real time, or obtain the data security protection measures in real time.

[0012] Furthermore, the reputation information of the participating entities is specifically to collect historical behavior records of the entities participating in data interaction, including the number of successful interactions, the number of failed interactions and violation records, to construct a reputation score, and the specific formula of the reputation score is as follows:

[0013]

[0014] Among them, S represents the number of successful interactions, F represents the number of failed interactions, V represents the number of violation records, and R represents the reputation score.

[0015] The metadata of data interactions between different entities is captured, where the data type of the metadata records the data type involved in each data interaction, such as text, image, video, and structured data. The data size records the amount of data in each data interaction, such as file size, number of records, etc. The interaction frequency records the frequency of data interaction, such as hourly, daily, weekly, etc. The reputation information of participating entities specifically collects historical behavior records of entities participating in data interactions, such as users, devices, and systems, including the number of successful interactions, the number of failed interactions, and violation records, to build a reputation score.

[0016] Furthermore, the dynamic deep learning model uses a custom encoder, and the specific implementation formula is as follows:

[0017] X'=f(W 2 ×σ(W 1 ×X+b 1 )+b 2 )

[0018] Among them, X represents the key feature vector of the input, W 1 and W 2 Represent different weight matrices, b 1 and b 2 They represent different bias items respectively, σ represents the activation function, such as the ReLU function, f represents the activation function of the output layer, such as the Sigmoid function, and X' represents the custom encoding output.

[0019] Furthermore, the data interaction uses real-time data analysis and behavior pattern recognition technology to detect abnormal behavior using an active security monitoring system, and once a security threat is discovered, preset security protection measures are immediately triggered.

[0020] The present invention uses an active supervision unit to continuously monitor data interaction activities in real time. Through real-time data analysis and behavior pattern recognition technology, the module can quickly detect and respond to any abnormal behavior. Once a potential security threat is discovered, it will immediately trigger the preset security protection measures to ensure the security of data interaction.

[0021] Furthermore, the dynamic deep learning model has an adaptive learning mechanism, which dynamically adjusts the deep learning model according to new data interaction patterns and security threats. The specific steps include:

[0022] Online learning: Learn from real-time data interactions and continuously update models to capture the latest security threats and data patterns;

[0023] Incremental training: Using incremental training methods, the model is fine-tuned to adapt to the latest data without retraining the entire model;

[0024] Model fusion: Combine the prediction results of dynamic deep learning models and use model fusion technology to improve the accuracy and robustness of risk assessment;

[0025] Feedback loop: Use feedback from security experts and the results of security incidents to train and optimize the model;

[0026] Multi-task learning: Improve the generalization ability of the model by learning multiple related tasks simultaneously through a multi-task learning framework.

[0027] The present invention also has an adaptive learning mechanism that can dynamically adjust the deep learning model according to new data interaction patterns and security threats to ensure the long-term effectiveness and adaptability of the system.

[0028] According to a second aspect of the present invention, a system for cross-subject data security risk assessment based on deep learning is proposed, which is characterized by comprising the following modules:

[0029] An intelligent data collection unit configured to capture metadata of data interactions between different entities, collect data types, scales and interaction frequencies of the metadata, and mine reputation information of participating entities, wherein the metadata and reputation information generate raw data;

[0030] An advanced feature construction unit configured to pre-process the raw data using an advanced feature builder, including data cleaning, normalization, and deduplication, wherein the advanced feature builder identifies and constructs key feature vectors related to data security risks through feature extraction technology;

[0031] A dynamic deep learning unit configured to input the key feature vector into a dynamic deep learning model, wherein the dynamic deep learning model continuously learns and optimizes model parameters through a back propagation algorithm;

[0032] The real-time risk assessment unit is configured to use the trained dynamic deep learning model to conduct in-depth analysis of the input key feature vectors, intelligently assess the security risk level of each data interaction, and output a quantitative risk probability value;

[0033] A strategy intelligent recommendation unit, configured to automatically formulate or recommend data security protection measures based on the risk probability value using a strategy intelligent recommendation system, wherein the strategy intelligent recommendation system includes data encryption, access permission control and data desensitization;

[0034] The interactive user operation unit is configured to use the interactive user operation platform to submit data interaction tasks through a web interface or a mobile application, view the security risk level in real time, or obtain the data security protection measures in real time.

[0035] According to a third aspect of the present invention, a computer program product is provided, on which one or more computer programs are stored. When the one or more computer programs are executed by a computer processor, the above method is implemented.

[0036] The above one or more technical solutions in the embodiments of the present application have at least one of the following technical effects:

[0037] 1) Improve the accuracy of risk identification: By leveraging the powerful feature extraction and pattern recognition capabilities of deep learning models, the system can more accurately identify potential security risks in cross-subject data interactions, including complex attack patterns and hidden security threats.

[0038] 2) Enhance the system's adaptive capabilities: The system continuously learns the patterns of cross-subject data interaction and environmental changes, automatically adjusts and optimizes risk assessment strategies without human intervention, and effectively responds to emerging security threats and changing data interaction scenarios.

[0039] 3) Automated and intelligent risk assessment: Through automated data feature analysis and risk prediction, the system reduces reliance on expert knowledge and human judgment, and improves the efficiency and intelligence of the assessment process.

[0040] 4) Strengthen real-time monitoring and rapid response: The system can monitor data interaction activities in real time, detect abnormal behavior in a timely manner, and quickly initiate early warning and response mechanisms to effectively reduce the risk of data leakage and abuse.

[0041] 5) Optimize user interaction and strategy generation: The user-friendly design of the user interface makes it easy for users to operate. At the same time, the system can intelligently generate or recommend customized data security protection strategies based on the evaluation results, thereby improving user participation and the effectiveness of strategy implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated into and constitute a part of this specification. The accompanying drawings illustrate the embodiments and are used together with the description to explain the principles of the present invention. It will be easy to recognize other embodiments and many expected advantages of the embodiments because they become better understood by reference to the following detailed description. The elements of the drawings are not necessarily to scale with each other. The same reference numerals refer to corresponding similar parts.

[0043] Figure 1 A schematic diagram of a method flow for cross-subject data security risk assessment based on deep learning according to an embodiment of the present invention is shown;

[0044] Figure 2 A schematic diagram of a system framework for cross-subject data security risk assessment based on deep learning according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0045] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0046] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0047] Figure 1 A schematic diagram of a method flow chart for cross-subject data security risk assessment based on deep learning according to an embodiment of the present invention is shown. Figure 1 As shown:

[0048] Capturing metadata of data interactions between different entities, collecting the data type, scale and interaction frequency of the metadata, and mining the reputation information of the participating entities, the metadata and reputation information generate raw data;

[0049] The reputation information of the participating entities is specifically to collect historical behavior records of the entities participating in data interaction, including the number of successful interactions, the number of failed interactions and violation records, to construct a reputation score. The specific formula of the reputation score is as follows:

[0050]

[0051] Among them, S represents the number of successful interactions, F represents the number of failed interactions, and V represents the number of violation records.

[0052] The present invention collects basic data types, scales and interaction frequencies, and also deeply mines the reputation information of participating entities to provide rich contextual information for subsequent risk assessment. Through flexible API integration, efficient web crawler technology or direct docking with data sources, the intelligent data collector can capture the required data in real time and accurately, laying a solid foundation for the in-depth analysis of the system.

[0053] Preprocessing the raw data using an advanced feature builder, including data cleaning, normalization, and deduplication, wherein the advanced feature builder identifies and constructs key feature vectors related to data security risks through feature extraction technology;

[0054] Based on data collection, the advanced feature builder performs a series of preprocessing operations on the raw data, including but not limited to data cleaning, normalization and deduplication, to ensure the quality and consistency of the data. In addition, through advanced feature extraction technology, key feature vectors closely related to data security risks are identified and constructed, such as data sensitivity, interaction frequency, and historical reputation of participating entities. These feature vectors will serve as inputs to the deep learning model to provide accurate data support for risk assessment.

[0055] Inputting the key feature vector into a dynamic deep learning model, wherein the dynamic deep learning model continuously learns and optimizes model parameters through a back propagation algorithm;

[0056] A variety of deep learning architectures are used, including convolutional neural networks (CNN), recurrent neural networks (RNN), and graph neural networks (GNN) to adapt to different types of data interaction patterns and security risk characteristics. By using historical data as a training set, the dynamic deep learning framework is able to continuously learn and optimize model parameters through the back-propagation algorithm to improve the accuracy of risk identification and the generalization ability of the model.

[0057] The trained dynamic deep learning model conducts in-depth analysis of the input key feature vectors, intelligently assesses the security risk level of each data interaction, and outputs a quantitative risk probability value;

[0058] The dynamic deep learning model uses a custom encoder, and the specific implementation formula is as follows:

[0059] X'=f(W 2 ×σ(W 1 ×X+b 1 )+b 2 )

[0060] Among them, X represents the key feature vector of the input, W 1 and W 2 Represent different weight matrices, b 1 and b 2 They represent different bias terms, σ represents the activation function, f represents the activation function of the output layer, and X' represents the custom encoding output.

[0061] The dynamic deep learning model has an adaptive learning mechanism, and dynamically adjusts the deep learning model according to new data interaction patterns and security threats. The specific steps include:

[0062] Online learning: Learn from real-time data interactions and continuously update models to capture the latest security threats and data patterns;

[0063] Incremental training: Using incremental training methods, the model is fine-tuned to adapt to the latest data without retraining the entire model;

[0064] Model fusion: Combine the prediction results of dynamic deep learning models and use model fusion technology to improve the accuracy and robustness of risk assessment;

[0065] Feedback loop: Use feedback from security experts and the results of security incidents to train and optimize the model;

[0066] Multi-task learning: Improve the generalization ability of the model by learning multiple related tasks simultaneously through a multi-task learning framework.

[0067] The data interaction uses real-time data analysis and behavior pattern recognition technology, and utilizes an active security monitoring system to detect abnormal behavior. Once a security threat is discovered, preset security protection measures are immediately triggered.

[0068] Based on the deep learning model, the real-time risk assessment core conducts in-depth analysis of the input feature vectors and intelligently assesses the security risk level of each data interaction. The core can not only provide qualitative risk assessment (such as high, medium, and low risk), but also give quantitative risk probability values, providing decision makers with more accurate risk measurement.

[0069] According to the risk probability value, a strategy intelligent recommendation system is used to automatically formulate or recommend data security protection measures, wherein the strategy intelligent recommendation system includes data encryption, access permission control and data desensitization;

[0070] Based on the output of the real-time risk assessment core, the policy intelligent recommendation system can automatically formulate or recommend the most appropriate data security protection measures. It has a rich built-in policy library, covering a variety of preset security policies such as data encryption, access control, data desensitization, etc., to ensure that effective protection measures can be provided for different risk levels.

[0071] Utilize the interactive user operation platform to submit data interaction tasks through the web interface or mobile application, view the security risk level in real time, or obtain the data security protection measures in real time.

[0072] In order to improve the user experience, the present invention provides an intuitive and highly interactive user operation platform. The platform supports users to easily submit data interaction tasks through a web interface or mobile application, view risk assessment results in real time, and obtain security policy recommendations. Users can enter detailed data interaction task information, and the system clearly displays the assessment results and security recommendations, allowing users to fully understand the security status of data interaction.

[0073] Based on the above dimensions, the present invention has developed a multi-dimensional security scoring system to evaluate the risks of data interaction from multiple angles and provide a more comprehensive security perspective. Specifically, it includes the following dimensions:

[0074] Data Sensitivity Score: Assess the sensitivity of data and the potential impact of a breach.

[0075] Entity Reputation Scoring: Assess the reputation of participating entities based on historical interactions and third-party data.

[0076] Interaction behavior scoring: Analyze data interaction behavior patterns and score abnormal or suspicious interactions.

[0077] Environmental risk score: Considers external environmental factors such as cyber attack trends, industry-specific threats, etc.

[0078] Compliance score: Evaluate whether data interaction complies with relevant regulations and policy requirements.

[0079] Comprehensive risk score: Combines the above scores to provide an overall risk assessment and gives risk mitigation suggestions.

[0080] In terms of security incident response, it can not only monitor and respond to security incidents in real time, but also provide incident response and data recovery mechanisms to minimize the impact of security incidents on users. Specific mechanisms include:

[0081] Instant alerts: When a security threat is detected, the system immediately alerts the security team and provides detailed information about the threat.

[0082] Automatic Isolation: The system can automatically isolate compromised data or system components to prevent the threat from spreading.

[0083] Forensic Analysis: Provides forensic analysis tools to help security experts determine the nature and source of threats.

[0084] Recovery strategy: Based on different types of security incidents, the system provides customized recovery strategies and steps.

[0085] Continuous Monitoring: After an incident is resolved, the system continues to monitor the affected systems and data to ensure that the threat is fully removed.

[0086] Reporting and documentation: The system generates detailed incident reports to record the incident handling process and results, providing data support for future security analysis and improvements.

[0087] In a specific embodiment, the functions are encapsulated into different modules to implement the method of the present invention, such as Figure 2 As shown in the figure, the system framework diagram of cross-subject data security risk assessment based on deep learning includes:

[0088] An intelligent data collection unit configured to capture metadata of data interactions between different entities, collect data types, scales and interaction frequencies of the metadata, and mine reputation information of participating entities, wherein the metadata and reputation information generate raw data;

[0089] An advanced feature construction unit configured to pre-process the raw data using an advanced feature builder, including data cleaning, normalization, and deduplication, wherein the advanced feature builder identifies and constructs key feature vectors related to data security risks through feature extraction technology;

[0090] A dynamic deep learning unit configured to input the key feature vector into a dynamic deep learning model, wherein the dynamic deep learning model continuously learns and optimizes model parameters through a back propagation algorithm;

[0091] The real-time risk assessment unit is configured to use the trained dynamic deep learning model to conduct in-depth analysis of the input key feature vectors, intelligently assess the security risk level of each data interaction, and output a quantitative risk probability value;

[0092] A strategy intelligent recommendation unit, configured to automatically formulate or recommend data security protection measures based on the risk probability value using a strategy intelligent recommendation system, wherein the strategy intelligent recommendation system includes data encryption, access permission control and data desensitization;

[0093] The interactive user operation unit is configured to use the interactive user operation platform to submit data interaction tasks through a web interface or a mobile application, view the security risk level in real time, or obtain the data security protection measures in real time.

[0094] In summary, the core technical features and advantageous technical effects of the present invention include:

[0095] 1) Innovative application of deep learning models: Deep learning models are used to conduct security risk assessments on cross-subject data interactions. The models can automatically extract features and learn complex patterns in data interactions to identify potential security threats.

[0096] 2) Automated feature engineering and risk assessment: An automated feature engineering process is implemented to convert raw data into feature vectors that are helpful for risk assessment, and deep learning models are used to perform accurate risk assessment.

[0097] 3) Real-time monitoring and intelligent response mechanism: The system has the ability to monitor cross-subject data interactions in real time, and can intelligently generate security strategies based on evaluation results and quickly respond to various security incidents.

[0098] 4) Intelligent user interface and policy generation: The present invention provides a user-friendly interface that allows users to operate conveniently, and the system can intelligently generate customized security policies based on the evaluation results, thereby enhancing the user experience and improving the efficiency of policy implementation.

[0099] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0100] The modules involved in the embodiments of the present application may be implemented by software or by hardware.

[0101] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A method for cross-subject data security risk assessment based on deep learning, characterized in that: The following steps are involved: Capturing metadata of data interactions between different entities, collecting the data type, scale and interaction frequency of the metadata, and mining the reputation information of the participating entities, the metadata and reputation information generate raw data; Preprocessing the raw data using an advanced feature builder, including data cleaning, normalization, and deduplication, wherein the advanced feature builder identifies and constructs key feature vectors related to data security risks through feature extraction technology; Inputting the key feature vector into a dynamic deep learning model, wherein the dynamic deep learning model continuously learns and optimizes model parameters through a back propagation algorithm; The trained dynamic deep learning model conducts in-depth analysis of the input key feature vectors, intelligently assesses the security risk level of each data interaction, and outputs a quantitative risk probability value; According to the risk probability value, a strategy intelligent recommendation system is used to automatically formulate or recommend data security protection measures, wherein the strategy intelligent recommendation system includes data encryption, access permission control and data desensitization; Utilize the interactive user operation platform to submit data interaction tasks through the web interface or mobile application, view the security risk level in real time, or obtain the data security protection measures in real time.

2. The method according to claim 1, characterized in that The reputation information of the participating entities is specifically to collect historical behavior records of the entities participating in data interaction, including the number of successful interactions, the number of failed interactions and violation records, to construct a reputation score. The specific formula of the reputation score is as follows: Among them, S represents the number of successful interactions, F represents the number of failed interactions, and V represents the number of violation records.

3. The method according to claim 1, characterized in that The dynamic deep learning model uses a custom encoder, and the specific implementation formula is as follows: X'=f(W2×σ(W1×X+b1)+b2) Among them, X represents the key feature vector of the input, W1 and W2 represent different weight matrices, b1 and b2 represent different bias terms, σ represents the activation function, f represents the activation function of the output layer, and X' represents the custom encoding output.

4. The method according to claim 1, characterized in that: The data interaction uses real-time data analysis and behavior pattern recognition technology, and utilizes an active security monitoring system to detect abnormal behavior. Once a security threat is discovered, preset security protection measures are immediately triggered.

5. The method according to claim 1 or 3, characterized in that: The dynamic deep learning model has an adaptive learning mechanism, and dynamically adjusts the deep learning model according to new data interaction patterns and security threats. The specific steps include: Online learning: Learn from real-time data interactions and continuously update models to capture the latest security threats and data patterns; Incremental training: Using incremental training methods, the model is fine-tuned to adapt to the latest data without retraining the entire model; Model fusion: Combine the prediction results of dynamic deep learning models and use model fusion technology to improve the accuracy and robustness of risk assessment; Feedback loop: Use feedback from security experts and the results of security incidents to train and optimize the model; Multi-task learning: Improve the generalization ability of the model by learning multiple related tasks simultaneously through a multi-task learning framework.

6. A system for cross-subject data security risk assessment based on deep learning, characterized in that: Includes the following modules: An intelligent data collection unit configured to capture metadata of data interactions between different entities, collect data types, scales and interaction frequencies of the metadata, and mine reputation information of participating entities, wherein the metadata and reputation information generate raw data; An advanced feature construction unit configured to pre-process the raw data using an advanced feature builder, including data cleaning, normalization, and deduplication, wherein the advanced feature builder identifies and constructs key feature vectors related to data security risks through feature extraction technology; A dynamic deep learning unit configured to input the key feature vector into a dynamic deep learning model, wherein the dynamic deep learning model continuously learns and optimizes model parameters through a back propagation algorithm; The real-time risk assessment unit is configured to use the trained dynamic deep learning model to conduct in-depth analysis of the input key feature vectors, intelligently assess the security risk level of each data interaction, and output a quantitative risk probability value; A strategy intelligent recommendation unit, configured to automatically formulate or recommend data security protection measures based on the risk probability value using a strategy intelligent recommendation system, wherein the strategy intelligent recommendation system includes data encryption, access permission control and data desensitization; The interactive user operation unit is configured to use the interactive user operation platform to submit data interaction tasks through a web interface or a mobile application, view the security risk level in real time, or obtain the data security protection measures in real time.

7. A computer program product, characterized in that A computer program is stored thereon, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

8. A computing system, characterized in that: The method comprises a processor and a memory, wherein the processor is configured to execute the method according to any one of claims 1 to 5.

Images