Method and system for optimizing data circulation in combination with AI (Artificial Intelligence) and trusted data space

Abstract

The invention provides a data circulation optimization method and system under AI combined trusted data space, and relates to the technical field of data communication, and the method comprises the steps: determining a data circulation demand, carrying out the intelligent matching of a data provider and a data user, and generating an optimal matching result; a trusted data space is constructed, data flow path optimization is carried out, and an optimal data flow path is generated; establishing a unified supervision platform, performing data circulation risk monitoring on each node, and outputting a risk index of each node; and when the risk index of each node is greater than a preset risk index, generating risk reminding information. According to the method and the device, the technical problem that the supervision demand of current data circulation is difficult to meet due to the problems of supervision blind areas, response lag and the like in the prior art can be solved, and the technical effects of effectively improving the safety, the efficiency and the intelligent level in the data circulation process and reducing potential risk events are achieved.

Description

Technical Field

[0001] The present application relates to the field of data communication technology, and in particular to a method and system for optimizing data flow in combination with AI in a trusted data space. Background Art

[0002] In the digital economy era, the circulation and application of data are of vital importance. However, with the increasing complexity of data circulation and the participation of multiple parties, how to ensure the security, compliance and efficiency of data circulation has become an urgent problem to be solved. Traditional data supervision methods often have problems such as blind spots and delayed responses, which are difficult to meet the current regulatory needs of data circulation. At the same time, establishing and maintaining trust between the multiple parties involved in the data circulation process (such as data providers, users, service providers, etc.) is also an important challenge.

[0003] To sum up, there are technical problems in the existing technology that make it difficult to meet the current regulatory needs for data circulation due to regulatory blind spots, delayed responses, and other issues. Summary of the invention

[0004] The purpose of this application is to provide a data flow optimization method and system in combination with AI in a trusted data space, in order to solve the technical problems mentioned above, that are difficult to meet the current regulatory needs for data flow due to the existence of regulatory blind spots, delayed responses, and other problems in the prior art.

[0005] In view of the above problems, the present application provides a method and system for optimizing data flow by combining AI with a trusted data space.

[0006] In the first aspect, the present application provides a data flow optimization method under AI combined with a trusted data space, and the data flow optimization method under AI combined with a trusted data space is implemented by an AI combined with a data flow optimization system under a trusted data space, wherein the data flow optimization method under AI combined with a trusted data space includes: determining data flow requirements, and intelligently matching data providers and data users based on the data flow requirements to generate optimal matching results; constructing a trusted data space, optimizing the data flow path in combination with the optimal matching results and the trusted data space, and generating an optimal data flow path; establishing a unified supervision platform based on each node in the optimal data flow path, performing data flow risk monitoring at each node, and outputting risk indicators of each node; when the risk indicator of each node is greater than a preset risk indicator, generating risk reminder information.

[0007] In the second aspect, the present application also provides an AI combined with a trusted data space data flow optimization system, which is used to execute the AI ​​combined with a trusted data space data flow optimization method as described in the first aspect, wherein the AI ​​combined with a trusted data space data flow optimization system includes: an intelligent matching module, used to determine the data flow demand, and based on the data flow demand, intelligently match the data provider and the data user to generate the optimal matching result; a data flow path optimization module, used to construct a trusted data space, and optimize the data flow path in combination with the optimal matching result and the trusted data space to generate the optimal data flow path; a risk monitoring module, used to establish a unified supervision platform based on each node in the optimal data flow path, perform data flow risk monitoring at each node, and output the risk index of each node; a risk reminder module, used to generate risk reminder information when the risk index of each node is greater than the preset risk index.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: Determine the data circulation demand, and intelligently match the data provider and the data user based on the data circulation demand to generate the optimal matching result; build a trusted data space, optimize the data circulation path in combination with the optimal matching result and the trusted data space, and generate the optimal data circulation path; establish a unified supervision platform based on each node in the optimal data circulation path, monitor the data circulation risk at each node, and output the risk index of each node; when the risk index of each node is greater than the preset risk index, generate risk reminder information. By combining artificial intelligence algorithms with trusted data spaces, intelligent analysis and matching of data circulation needs can be achieved, data circulation paths can be optimized, and risks of each node can be monitored in real time, so as to effectively improve the security, efficiency and intelligence level of the data circulation process and reduce the technical effect of potential risk events.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 This is a flowchart of the data flow optimization method in this application combining AI with a trusted data space.

[0012] Figure 2 This is a structural diagram of the data flow optimization system combined with AI in the trusted data space in this application.

[0013] Explanation of the accompanying drawings: intelligent matching module 11, data flow path optimization module 12, risk monitoring module 13, risk reminder module 14. DETAILED DESCRIPTION

[0014] This application solves the technical problems mentioned above that exist in the prior art, such as the difficulty in meeting the current regulatory requirements for data circulation due to the existence of regulatory blind spots and delayed responses, by providing a data circulation optimization method and system in combination with AI in a trusted data space. By combining artificial intelligence algorithms with trusted data spaces, intelligent analysis and matching of data circulation requirements can be achieved, data circulation paths can be optimized, and risks at each node can be monitored in real time, so as to effectively improve the security, efficiency, and intelligence level of the data circulation process and reduce potential risk events.

[0015] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0016] For example, please refer to the attached Figure 1 The present application provides a data flow optimization method in combination with AI in a trusted data space, wherein the data flow optimization method in combination with AI in a trusted data space is applied to a data flow optimization system in combination with AI in a trusted data space, and the data flow optimization method in combination with AI in a trusted data space specifically includes the following steps: Step 1: Determine the data circulation demand, and intelligently match the data provider and the data user based on the data circulation demand to generate the optimal matching result.

[0017] Specifically, in the process of data circulation, it is first necessary to clearly define and analyze the data circulation needs. Data circulation needs usually include requirements for data type, data volume, transmission timeliness, and data quality. In order to meet these requirements, the characteristics of data providers and data users must be accurately identified and analyzed. Data providers usually refer to entities that own data and are willing to share or sell data, while data users are entities that need to obtain data for analysis, decision-making or application. Based on a deep understanding of these needs, artificial intelligence algorithms are used for intelligent matching to ensure that the needs between providers and users are matched to the greatest extent possible, thereby optimizing the entire data circulation process.

[0018] The process of intelligent matching first requires feature extraction of data providers and data users. For data providers, these features may include factors such as the type of their data (such as text, pictures, videos, etc.), the quality of the data (such as accuracy, completeness, etc.), the frequency of data updates, and the credibility of the provider; for data users, feature extraction may include specific requirements for the data they need (such as data in a specific industry, data in a specific time period, etc.), data format requirements, data accuracy requirements, etc. Through these feature extractions, a demand feature model for data providers and data users can be constructed. After feature extraction, the demand features between data providers and data users are intelligently matched. The optimal matching results generated will become the basis for optimizing the data circulation path, providing an important reference for the subsequent construction of a trusted data space and the selection of data circulation paths. For example, in a data circulation scenario in the financial industry, through intelligent matching, financial institutions can find the most suitable data providers and users under different regulatory environments, ensuring that data circulation meets financial regulatory requirements while meeting data timeliness and quality requirements.

[0019] Step 2: Construct a trusted data space, optimize the data flow path by combining the optimal matching result and the trusted data space, and generate an optimal data flow path.

[0020] Specifically, building a trusted data space and optimizing the data flow path based on the optimal matching results are key steps to ensure efficient and secure data flow. In this process, the construction of a trusted data space provides a safe and transparent environment for data flow, while data flow path optimization selects the best data transmission route based on the optimal matching results to ensure a balance between cost, efficiency and security in the flow process.

[0021] First of all, a trusted data space is an environment that ensures the security, compliance and transparency of data circulation through various mechanisms. The core of building a trusted data space is to provide a unified framework to ensure that the source, transmission, storage and use of data can be effectively supervised and audited at every link of data circulation. This space usually includes multiple trust nodes (such as data providers, data users, regulatory agencies, etc.), and each node is given clear authority and responsibilities. In order to ensure the authenticity and credibility of the data, encryption technology, digital signatures, data quality control and other means are usually used to prevent data from being tampered with during transmission.

[0022] When building a trusted data space, it is necessary to select a suitable data flow path based on the optimal matching result, that is, the optimal data flow path, and ensure the security and compliance of each link in the path through the security measures of the trusted data space. Ultimately, the flow of data from the provider to the user is not only more efficient, but also ensures the transparency, traceability and security of the entire process.

[0023] Step three: Based on each node in the optimal data circulation path, a unified supervision platform is established to monitor the data circulation risks at each node and output the risk indicators of each node.

[0024] Specifically, in the process of optimizing the data circulation path, establishing a unified supervision platform for data circulation risk monitoring is a key link to ensure the safety and compliance of data circulation. By conducting real-time risk monitoring at each node, potential safety hazards, compliance issues or operational errors can be discovered in a timely manner, thereby reducing the risk of data leakage, tampering or improper use. The unified supervision platform will act as a centralized data monitoring and risk assessment system, capable of real-time tracking of all nodes in the data circulation path and assessing the risk level of each node in the data circulation process. The unified supervision platform first needs to collect access record data for each node, including data request, data transmission, data storage, access rights, operation logs and other information. These records will help the platform track the flow of data at each node, determine whether the data is processed according to the prescribed process, and whether there are any abnormal operations. For example, if a data provider frequently accesses sensitive data of other nodes, or if the access rights of a data user exceed the scope of authorization, they can be discovered in time through access records.

[0025] After obtaining the access records of each node, the behavior of each node is evaluated based on the preset data circulation risk prediction model, and the risk index of each node is generated according to the access records and operation behaviors of the node. Through this risk monitoring mechanism, potential security risks can be identified and warned in real time during the data circulation process, ensuring the security, integrity and compliance of data in the entire transmission link, and minimizing the risk of data leakage, tampering or improper use.

[0026] Step 4: When the risk index of each node is greater than the preset risk index, risk warning information is generated.

[0027] Specifically, in the risk management of the data circulation process, once the node risk index generated by the data circulation risk prediction model and similarity comparison exceeds the preset risk index, the risk reminder information will be automatically triggered to remind the relevant managers or system administrators to take necessary measures. This mechanism can ensure that when the risk level is too high, timely response measures are taken to prevent potential security incidents or compliance issues. Risk reminder information can be notified to relevant personnel in a variety of ways, such as SMS notification, email notification, system message, mobile application push, etc.

[0028] Furthermore, step one of this application includes: Based on the data circulation demand, the characteristics of the data provider are extracted to generate the provider demand characteristics; based on the data circulation demand, the characteristics of the data user are extracted to generate the user demand characteristics; the provider demand characteristics and the user demand characteristics are matched to establish the optimal matching result.

[0029] Specifically, in the process of optimizing data circulation, it is first necessary to extract features from data providers and data users based on the specific needs of data circulation. Feature extraction is to quantify and describe the needs and capabilities of each party so that accurate matching can be performed during the matching process. For feature extraction of data providers, multiple dimensions are usually considered, such as the data type of the data provider (such as structured data, unstructured data, time series data, etc.), data quality (such as data accuracy, completeness, update frequency, etc.), data availability (such as data timeliness, accessibility, etc.). In addition, the processing capabilities of the data provider need to be evaluated, that is, whether it can cope with large-scale data needs, or whether it can provide customized data formats. These features can be extracted through analysis of the provider's historical data circulation records, evaluation of the provider's technical architecture capabilities, and compliance checks of industry standards, so as to obtain the provider's demand features. Feature extraction of data users focuses on their needs for data, such as the type of data needs of the user (such as whether real-time data, batch data, etc. are required), data format requirements (such as JSON, CSV format, etc.), data accuracy (such as high-precision sensor data or roughly estimated data), and the purpose of data use (such as analysis, prediction, reporting, etc.). In addition, users may also have compliance requirements, such as whether data needs to meet certain privacy protection or data security standards. Through these features, the user's specific requirements for data can be clarified, including their trust in the data source, the demand for data update frequency, etc., so as to obtain the user's demand characteristics.

[0030] After the feature extraction of the provider and the user is completed, the next step is feature matching. Through similarity analysis, the provider's demand features and the user's demand features can be compared to determine which pair of provider and user demand features are the most matched, obtain the best matching result, and determine the most suitable combination of provider and user, thereby achieving an efficient, compliant, and secure optimization path for data circulation, which not only improves the efficiency of data circulation, but also ensures a high degree of fit between data quality and demand.

[0031] Furthermore, the present application also includes the following steps: A similarity analysis is performed on the provider's demand characteristics and the user's demand characteristics to generate a number of matching degrees corresponding to a number of matching results; based on the number of matching degrees, a number of matching results with similarities greater than or equal to a preset similarity are extracted to generate the optimal matching result.

[0032] Specifically, in the process of optimizing data circulation, similarity analysis between the demand characteristics of providers and users is a crucial step. By conducting quantitative matching analysis on these characteristics, we can ensure that the needs of the selected data providers and users are as consistent as possible, thereby achieving the best data circulation effect.

[0033] First, in order to perform similarity analysis, it is necessary to clarify the feature vectors of the data provider and the data user, and extract the feature vectors based on the provider's demand characteristics and the user's demand characteristics. It is assumed that the demand characteristics of each provider and user can be represented by a multi-dimensional vector, and each dimension of the vector represents a different demand attribute, such as data type, data quality, update frequency, accuracy requirements, etc. For example, the demand feature vector of data provider A may be: [data type = time series data, data quality = high, update frequency = hourly, data format = JSON], while the demand feature vector of data user B may be: [data type = time series data, data quality = medium, update frequency = hourly, data format = CSV]. Through such feature representation, a corresponding feature vector can be generated for each provider and user.

[0034] Next, similarity analysis is used to calculate the matching degree between the data provider and the data user. Common similarity analysis methods include cosine similarity, Euclidean distance, etc., to obtain a matching value. For example, assuming that in the experiment, the matching degree between provider A and user B is 0.85, the matching degree between provider C and user D is 0.92, and the matching degree between provider E and user F is 0.78. These values ​​represent the adaptability between different matching pairs. Based on these similarity values, screening is performed. In order to ensure that the selected matching results meet certain quality requirements, a preset similarity threshold can be set. Only those matching results with a matching degree greater than or equal to the threshold will be considered valid matches. For example, assuming that the preset similarity threshold is 0.80, then the matching results with a matching degree of 0.85 and 0.92 will be considered valid, while the matching result with a matching degree of 0.78 will not meet the requirements. Through this screening process, matching pairs with a similarity higher than the threshold are extracted from all possible matching results. These selected matching pairs will constitute the optimal matching results and become the basis for subsequent data flow path optimization.

[0035] To sum up, by analyzing the similarity of the demand characteristics of providers and users and screening out the optimal matching results based on the matching degree, not only the efficiency of data circulation is improved, but also the foundation is laid for the subsequent optimization of data circulation paths, ensuring the security, compliance and efficiency of the data circulation process.

[0036] Further, step 2 of this application includes: A data circulation network is constructed based on the optimal matching result, wherein the data circulation network includes multiple circulation nodes, and any node has identification information identifying a data provider or a data user; the trusted data space is constructed based on the multiple circulation nodes; multiple data circulation paths are generated based on the optimal matching result and the trusted data space; a multi-objective optimization model is called to optimize the multiple data circulation paths to generate the optimal data circulation path.

[0037] Specifically, in the process of data circulation optimization, it is first necessary to build a data circulation network based on the optimal matching results. The key to this process is to use data providers and data users as nodes in the network and assign identification information to each node. These nodes represent data providers and demanders, and the paths connecting these nodes constitute the routes of data circulation. When building a data circulation network, each node not only contains the basic identification information of the data provider or data user, but also includes the relevant attributes of the node, including data type, data quality, data update frequency, data format and other information. For example, the node of data provider A may be identified as provider A, and accompanied by information such as data type, quality level and update time; similarly, the node of data user B is identified as user B, and accompanied by the type and format requirements of the data required. Once the nodes and attributes of the data circulation network are identified, the next step is to build a trusted data space based on these nodes. The trusted data space is not just a network containing nodes and connections. It also includes reputation assessment, compliance monitoring, data protection measures, etc. for each node, ensuring that in the process of data circulation, each data transmission link can be fully monitored and audited. For example, the trusted data space may set rules requiring all data providers to pass a certain authentication process to become nodes in the network, and all data transmission must use encryption technology to ensure data security.

[0038] With a trusted data space, multiple data flow paths can be generated in this space based on the optimal matching results. Each path represents a different transmission path for data from the provider to the user, which may involve different transit nodes, different transmission protocols, and different data processing methods. In order to select the best flow path among these paths, the multi-objective optimization model will be called next. The role of the multi-objective optimization model is to balance the conflicts and trade-offs between multiple goals. In the optimization of data flow paths, common optimization goals include: transmission cost (how to achieve data circulation at a lower cost), transmission efficiency (how to increase data transmission speed and reduce delays), and data transmission security (how to ensure the security of data during transmission and prevent data leakage or tampering). Through multi-objective optimization, each data flow path can be evaluated, considering the weight of each goal, and finally selecting the path that best meets the needs, that is, the optimal data flow path, to complete data transmission at the best cost and efficiency while ensuring data security. Ultimately, this optimized data flow path provides an efficient, secure, and compliant solution for data flow between data providers and users. It not only improves the use value of data, but also ensures the transparency and traceability of the data flow process, thereby further enhancing the credibility and stability of the entire data flow ecosystem.

[0039] Furthermore, the present application also includes the following steps: The multi-objective optimization model is used to evaluate the data transmission performance of the multiple data flow paths in multiple dimensions to generate multiple evaluation indicators, wherein the multiple dimensions include at least data transmission cost, data transmission efficiency and data transmission security; based on the multiple evaluation indicators, the data flow path corresponding to the optimal evaluation indicator is selected as the optimal data flow path.

[0040] Specifically, in the process of optimizing data flow paths, multi-objective optimization models play a vital role. They can evaluate different flow paths according to multiple dimensions and finally select the optimal path. This model will comprehensively consider multiple performance indicators of data transmission, including key factors such as data transmission cost, efficiency and security, to ensure that in practical applications, the selected path can not only meet business needs, but also ensure the efficiency and security of the data flow process. First of all, data transmission cost is an important dimension for evaluating data flow paths. Data transmission cost usually includes physical transmission cost (such as bandwidth fee, network usage fee), equipment and facility maintenance cost, and management and monitoring cost. For a typical data flow path, different transmission methods (such as direct transmission, transmission through multiple transit nodes) will involve different costs. The multi-objective optimization model calculates the comprehensive cost of each path and helps determine which paths are more cost-effective by comparing the costs of different paths. For example, in some cases, choosing a path may require passing through multiple transit nodes, which will increase bandwidth costs and management costs, while choosing another direct transmission path may be able to reduce costs, but in some cases, it may sacrifice transmission efficiency or security.

[0041] Secondly, data transmission efficiency is another key optimization goal, which usually includes factors such as data transmission speed and delay. The transmission efficiency directly affects the data transmission time, and then affects the timeliness of data delivery. In the evaluation of data flow paths, the transmission speed may be determined by factors such as network bandwidth, network quality, and routing complexity, while the delay is closely related to factors such as the number of transit nodes and network congestion. For example, in real-time data transmission scenarios, data transmission speed is crucial, and any delay may affect the timeliness of data and the decision-making efficiency of users. Therefore, in multi-objective optimization, paths with higher efficiency will be given priority to ensure that data can reach users at the fastest speed.

[0042] Third, data transmission security is a key factor in ensuring the confidentiality, integrity and availability of data during transmission. In an environment of trusted data circulation, ensuring data security is not only about preventing data loss, but also about preventing data tampering.

[0043] Through the comprehensive evaluation of these dimensions, multiple evaluation indicators will be generated, which respectively represent the performance of each path in terms of cost, efficiency and safety. The evaluation values ​​of each dimension can be integrated by weighted average, hierarchical analysis or other mathematical methods. For example, suppose that path 1 is scored 80 points in cost, 90 points in efficiency, and 85 points in safety; path 2 is scored 70 points in cost, 80 points in efficiency, and 95 points in safety. In this case, the comprehensive score of each path may be calculated according to the preset weights (such as 40% for cost, 30% for efficiency, and 30% for safety), and the path with the highest score is finally selected as the optimal path. For example, suppose there are three candidate paths. After evaluation by the multi-objective optimization model, the following evaluation results are obtained: Path 1: cost score 85, efficiency score 90, safety score 80, and comprehensive score 85. Path 2: cost score 80, efficiency score 85, safety score 90, and comprehensive score 85.5. Path 3: cost score 90, efficiency score 75, safety score 85, and comprehensive score 82.5. In this example, path 2 has the highest comprehensive score, which means that it has the best comprehensive performance in the three dimensions of cost, efficiency and security. Therefore, path 2 will be selected as the optimal data flow path.

[0044] Through the multi-objective optimization model, the selection of data flow path no longer depends solely on a certain dimension, but through comprehensive consideration of multiple key factors, it ensures that the selected path is efficient, secure and cost-effective in practical applications. This optimization process is particularly important for various scenarios that require efficient data exchange, such as real-time transaction data transmission in the financial industry, medical record data sharing in the medical industry, etc., which can ensure the rapid and secure flow of data while reducing the waste of resources.

[0045] Further, step three of this application includes: Node access records of each node are collected through the unified supervision platform to generate multiple node access record data sets; based on the multiple node access record data sets, data circulation risk prediction models are used to monitor data circulation risks and output risk indicators of each node.

[0046] Specifically, in the process of optimizing data circulation, one of the core tasks of the unified supervision platform is to collect and analyze the access records of each node in real time in order to generate evaluation indicators for data circulation risks. By establishing a node access record data set and conducting in-depth analysis of this data with the help of a data circulation risk prediction model, it is possible to effectively monitor the security of each node during the data circulation process, discover potential risks in a timely manner, and provide effective early warning and decision support.

[0047] First, the unified supervision platform needs to collect detailed node access records from each node (including data providers and data users), including the access behavior of each node, the time and frequency of access, the type of data accessed, the source of the access request, the content of data access, the data operation behavior, etc., to obtain multiple node access record data sets. After establishing the access record data sets of multiple nodes, the next step is to analyze these data using the data circulation risk prediction model to assess the risk level of each node. The data circulation risk prediction model is usually based on historical data, machine learning or deep learning algorithms, combined with the node's access behavior pattern, known security risks and compliance requirements, to predict the risks that may arise in the node in the future, and generate a comprehensive risk score for each node, that is, a risk index. The higher the score, the greater the potential risk of the node. Through this data circulation risk prediction and monitoring based on node access records, the unified supervision platform can evaluate the risk status of each node in real time and dynamically, discover potential safety hazards in a timely manner, and maximize the security, compliance and stability of data circulation through automated early warnings and measures.

[0048] Furthermore, the present application also includes the following steps: The risk sample database is called through the data circulation risk prediction model; each access record sample in the risk sample database is traversed to perform similarity comparison with the multiple node access record data sets, and the data circulation risk samples corresponding to the access record samples with similarity greater than the preset similarity are obtained for output to generate the risk index of each node.

[0049] Specifically, the risk sample database is a large database that contains historical data circulation events and known risk patterns. It stores various risk events that have occurred, including abuse of authority, data leakage, malicious tampering, compliance violations, etc. These historical event samples can be used as basic data for training models and as reference samples for predicting current node risks. Each risk sample contains an access record sample and a corresponding data circulation risk sample. The node access record data set of each node is compared with each access record sample in the risk sample database. The core of the comparison is through similarity analysis, that is, the behavioral characteristics of the current node are compared with the characteristics of historical risk events, and the similarity value between them is calculated. Common similarity calculation methods include: cosine similarity: measure the angle between two sample vectors in the feature space. The smaller the angle, the higher the similarity; Euclidean distance: by calculating the Euclidean distance between two sample feature vectors, the smaller the distance, the higher the similarity.

[0050] A similarity value will be obtained between each access record sample and the historical risk event sample. According to the preset similarity, the historical risk event samples with similarity greater than or equal to the threshold are screened out. In this way, the data circulation risk samples corresponding to the access record samples that are most similar to the access record data sets of multiple nodes can be identified and output as the risk indicators of each node. By combining the risk sample database with similarity analysis, the unified supervision platform can dynamically monitor the behavior of data circulation nodes based on historical risk samples, timely discover potential risks, and generate targeted risk indicators. In this way, not only can the sensitivity to abnormal behavior be improved, but also personalized security monitoring and early warning services can be provided for each node in the data circulation process, thereby ensuring the security and compliance of data circulation to the greatest extent.

[0051] In summary, the data flow optimization method provided by this application in combination with AI in a trusted data space has the following technical effects: Determine the data circulation demand, and intelligently match the data provider and the data user based on the data circulation demand to generate the optimal matching result; build a trusted data space, optimize the data circulation path in combination with the optimal matching result and the trusted data space, and generate the optimal data circulation path; establish a unified supervision platform based on each node in the optimal data circulation path, monitor the data circulation risk at each node, and output the risk index of each node; when the risk index of each node is greater than the preset risk index, generate risk reminder information. By combining artificial intelligence algorithms with trusted data spaces, intelligent analysis and matching of data circulation needs can be achieved, data circulation paths can be optimized, and risks of each node can be monitored in real time, so as to effectively improve the security, efficiency and intelligence level of the data circulation process and reduce the technical effect of potential risk events.

[0052] Embodiment 2: Based on the same inventive concept as the data flow optimization method in the aforementioned embodiment in combination with AI in a trusted data space, this application also provides a data flow optimization system in combination with AI in a trusted data space, please refer to the attached Figure 2 The data flow optimization system of AI combined with the trusted data space includes: The intelligent matching module 11 is used to determine the data circulation demand, and based on the data circulation demand, intelligently match the data provider and the data user to generate the best matching result.

[0053] The data flow path optimization module 12 is used to construct a trusted data space, optimize the data flow path in combination with the optimal matching result and the trusted data space, and generate an optimal data flow path.

[0054] The risk monitoring module 13 is used to establish a unified supervision platform based on each node in the optimal data circulation path, perform data circulation risk monitoring at each node, and output the risk index of each node.

[0055] The risk reminder module 14 is used to generate risk reminder information when the risk index of each node is greater than a preset risk index.

[0056] Furthermore, the intelligent matching module 11 in the data flow optimization system in the AI ​​combined with the trusted data space is also used for: Based on the data circulation demand, the characteristics of the data provider are extracted to generate the provider demand characteristics; based on the data circulation demand, the characteristics of the data user are extracted to generate the user demand characteristics; the provider demand characteristics and the user demand characteristics are matched to establish the optimal matching result.

[0057] Furthermore, the intelligent matching module 11 in the data flow optimization system in the AI ​​combined with the trusted data space is also used for: A similarity analysis is performed on the provider's demand characteristics and the user's demand characteristics to generate a number of matching degrees corresponding to a number of matching results; based on the number of matching degrees, a number of matching results with similarities greater than or equal to a preset similarity are extracted to generate the optimal matching result.

[0058] Furthermore, the data flow path optimization module 12 in the data flow optimization system in the AI ​​combined with the trusted data space is also used for: A data circulation network is constructed based on the optimal matching result, wherein the data circulation network includes multiple circulation nodes, and any node has identification information identifying a data provider or a data user; the trusted data space is constructed based on the multiple circulation nodes; multiple data circulation paths are generated based on the optimal matching result and the trusted data space; a multi-objective optimization model is called to optimize the multiple data circulation paths to generate the optimal data circulation path.

[0059] Furthermore, the data flow path optimization module 12 in the data flow optimization system in the AI ​​combined with the trusted data space is also used for: The multi-objective optimization model is used to evaluate the data transmission performance of the multiple data flow paths in multiple dimensions to generate multiple evaluation indicators, wherein the multiple dimensions include at least data transmission cost, data transmission efficiency and data transmission security; based on the multiple evaluation indicators, the data flow path corresponding to the optimal evaluation indicator is selected as the optimal data flow path.

[0060] Furthermore, the risk monitoring module 13 in the data flow optimization system in the AI ​​combined with the trusted data space is also used for: Node access records of each node are collected through the unified supervision platform to generate multiple node access record data sets; based on the multiple node access record data sets, data circulation risk prediction models are used to monitor data circulation risks and output risk indicators of each node.

[0061] Furthermore, the risk monitoring module 13 in the data flow optimization system in the AI ​​combined with the trusted data space is also used for: The risk sample database is called through the data circulation risk prediction model; each access record sample in the risk sample database is traversed to perform similarity comparison with the multiple node access record data sets, and the data circulation risk samples corresponding to the access record samples with similarity greater than the preset similarity are obtained for output to generate the risk index of each node.

[0062] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The data flow optimization method and specific examples of AI combined with trusted data space in Example 1 are also applicable to the data flow optimization system of AI combined with trusted data space in this embodiment. Through the detailed description of the data flow optimization method of AI combined with trusted data space, those skilled in the art can clearly know the data flow optimization system of AI combined with trusted data space in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0063] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0064] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. The data flow optimization method in AI combined with trusted data space is characterized by: include: Determine data circulation requirements, and intelligently match data providers and data users based on the data circulation requirements to generate optimal matching results; Constructing a trusted data space, optimizing a data flow path by combining the optimal matching result and the trusted data space, and generating an optimal data flow path; Based on each node in the optimal data circulation path, a unified supervision platform is established to monitor the data circulation risk at each node and output the risk index of each node; When the risk index of each node is greater than the preset risk index, risk reminder information is generated.

2. The data flow optimization method in AI combined with trusted data space as claimed in claim 1 is characterized in that: Based on the data circulation requirements, intelligent matching is performed between data providers and data users, and the optimal matching results are generated, including: Extracting characteristics of data providers based on the data circulation requirements to generate provider requirement characteristics; Extracting characteristics of data users based on the data circulation requirements to generate user demand characteristics; The provider's demand characteristics and the user's demand characteristics are matched to establish the optimal matching result.

3. The data flow optimization method in AI combined with trusted data space as claimed in claim 2 is characterized in that: Matching the provider's demand characteristics with the user's demand characteristics to establish the optimal matching result includes: Performing similarity analysis on the provider's demand characteristics and the user's demand characteristics to generate a plurality of matching degrees corresponding to a plurality of matching results; Based on the plurality of matching degrees, a plurality of matching results having a similarity greater than or equal to a preset similarity are extracted to generate the optimal matching result.

4. The data flow optimization method in AI combined with trusted data space as claimed in claim 1, characterized in that: Constructing a trusted data space, optimizing the data flow path by combining the optimal matching result and the trusted data space, and generating an optimal data flow path, including: Building a data circulation network based on the optimal matching result, wherein the data circulation network includes a plurality of circulation nodes, and any node has identification information identifying a data provider or a data user; Building the trusted data space based on the multiple circulation nodes; generating a plurality of data flow paths based on the optimal matching result and the trusted data space; The multi-objective optimization model is called to optimize the multiple data flow paths to generate the optimal data flow path.

5. The data flow optimization method in AI combined with trusted data space as claimed in claim 4 is characterized in that: Calling a multi-objective optimization model to optimize the multiple data flow paths to generate the optimal data flow path includes: The multi-objective optimization model is used to evaluate the data transmission performance of the multiple data flow paths under multiple dimensions to generate multiple evaluation indicators, wherein the multiple dimensions at least include data transmission cost, data transmission efficiency and data transmission security; Based on the multiple evaluation indicators, a data flow path corresponding to an optimal evaluation indicator is selected as the optimal data flow path.

6. The data flow optimization method in AI combined with trusted data space as claimed in claim 1, characterized in that: Based on each node in the optimal data flow path, a unified supervision platform is established to monitor data flow risks at each node and output risk indicators of each node, including: Collect node access records of each node through the unified supervision platform to generate multiple node access record data sets; Based on the multiple node access record data sets, data circulation risk monitoring is performed using a data circulation risk prediction model to output risk indicators for each node.

7. The data flow optimization method in AI combined with trusted data space as claimed in claim 6 is characterized in that: Based on the plurality of node access record data sets, data circulation risk monitoring is performed using a data circulation risk prediction model, and risk indicators of each node are output, including: Calling the risk sample database through the data circulation risk prediction model; Each access record sample in the risk sample database is traversed to perform similarity comparison with the multiple node access record data sets, and data circulation risk samples corresponding to the access record samples with similarity greater than a preset similarity are obtained for output to generate risk indicators for each node.

8. AI combined with the data flow optimization system in the trusted data space, characterized by: Steps for implementing the method for optimizing data flow in a data space in combination with AI in any one of claims 1 to 7, wherein the system for optimizing data flow in a data space in combination with AI in combination with trusted data space comprises: An intelligent matching module is used to determine data circulation requirements, and based on the data circulation requirements, intelligently match data providers with data users to generate optimal matching results; A data flow path optimization module, used to construct a trusted data space, optimize the data flow path by combining the optimal matching result and the trusted data space, and generate an optimal data flow path; A risk monitoring module, used to establish a unified supervision platform based on each node in the optimal data circulation path, perform data circulation risk monitoring on each node, and output risk indicators of each node; The risk reminder module is used to generate risk reminder information when the risk index of each node is greater than the preset risk index.

Images