Information management system of government affair information system based on data sharing and exchange

Abstract

The invention belongs to the technical field of government information management, and discloses an information management system of a government information system based on data sharing and exchange. The system comprises a data processing end, an information supervision end and an information sharing end, the data processing end is used for realizing standardized processing aiming at the problems of missing, redundancy and format disorder in government affair data, and monitoring the quality state of the government affair information data in the whole process of collection, transmission and storage in real time; the information supervision end is used for establishing a credibility quantification mechanism, providing a basis for shared authorization and risk management and control, adjusting access permission in real time, and auditing access behaviors to prevent unauthorized access; and the information sharing end is used for realizing cross-department data collaborative calculation on the premise of not leaking original data, ensuring privacy security, and realizing multi-department data association fusion and dynamic permission allocation at the same time. The information management system of the government affair information system based on data sharing and exchange successfully realizes integrated closed-loop operation management of data receiving, management and sharing.

Description

Technical Field

[0001] This invention relates to the field of government information management technology, and more specifically, to an information management system based on a government information system for data sharing and exchange. Background Technology

[0002] Government information systems based on data sharing and exchange are the core support for breaking down departmental data barriers and achieving cross-domain collaboration. They can integrate scattered government data resources, improve the efficiency of government services, and support cross-departmental business collaboration, making them a key infrastructure for the construction of digital government.

[0003] In terms of data governance, existing technologies mostly involve static processing for cleaning and quality monitoring, which is insufficient for adapting to dynamic data. They also lack a systematic quantitative mechanism for trust assessment, affecting the accuracy of sharing authorization. Furthermore, access control has limited flexibility, making it difficult to dynamically balance sharing efficiency and security needs. Privacy protection measures are relatively simplistic, and there is a risk of information leakage when conducting cross-departmental collaborative computing. Data fusion often relies on a single key field, resulting in insufficient correlation accuracy, which to some extent restricts the full release of the value of data sharing.

[0004] In view of this, the present invention proposes an information management system for government information systems based on data sharing and exchange to solve the above problems. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an information management system for a government information system based on data sharing and exchange includes: a data processing terminal, an information supervision terminal, and an information sharing terminal;

[0006] The data processing terminal includes a data cleaning module and a data quality monitoring module;

[0007] The data cleaning module includes a missing value intelligent repair unit and a redundant data removal unit;

[0008] The intelligent missing value repair unit prioritizes government information data based on its importance. It then filters samples from the same batch of data that do not contain missing values ​​for the target missing field, excluding obvious outliers. For the filtered non-missing samples, it assigns corresponding weights based on the priority of their respective fields. The non-missing sample values ​​and their corresponding weights are then substituted into a formula to calculate the repair value. The specific calculation formula is as follows:

[0009] ;

[0010] Get repair value ,in, Let i be the field weight of the i-th non-missing sample. Let i be the target field value of the i-th non-missing sample. The total number of valid non-missing samples is used to calculate the repair value. The purpose of this calculation is to distinguish the importance of government affairs fields by weight, ensure the repair accuracy of core fields, and avoid cross-departmental data matching failures due to repair errors.

[0011] The redundant data removal unit selects key fields based on the data type of government affairs, and divides the key fields into text-type and numerical-type. The similarity of text-type fields is calculated by word vector cosine, and the similarity of numerical-type fields is calculated by relative error. At the same time, the similarity of a single field is weighted and summed with its corresponding weight to calculate the comprehensive similarity. When the comprehensive similarity is greater than or equal to the redundancy threshold, it is determined to be redundant data. The most complete segment of the redundant data is retained, and the remaining redundant data is marked as data to be deleted and sent to the data management module for manual review and deletion.

[0012] The information monitoring terminal includes a trust assessment module and a dynamic access control module;

[0013] The data quality monitoring module includes a data integrity monitoring unit and a data accuracy dynamic verification unit;

[0014] The data integrity monitoring unit, based on government data standards, calculates field missing rate and record integrity score to monitor whether the data meets the basic requirements for sharing.

[0015] Furthermore, based on government data standards, the steps for calculating field missing rates and record integrity scores to monitor whether data meets basic sharing requirements include:

[0016] Step 1: Based on government data standards, determine the required and optional fields for each type of data; for each batch of data access, count the number of missing records for each field and calculate the field missing rate;

[0017] Step Two: Assign high weight to required fields and low weight to optional fields to ensure that the absence of core fields has a greater impact on completeness. Calculate the completeness score by substituting the values ​​into the formula. The specific calculation formula is as follows:

[0018] ;

[0019] Obtain integrity score ,in, This is the total number of fields required by the government data standard. Let k be the missing rate of the k-th field. The weight of the k-th field is used to calculate the integrity score to quantitatively assess data integrity, avoid the inability of shared data to be used due to the missing key fields, and balance monitoring accuracy and collection efficiency through weight.

[0020] Step 3: Based on the integrity score calculated in Step 2, a judgment is made. When the integrity score is less than the score threshold, a repair instruction is pushed to the data provider, specifying the missing fields and the time limit for filling them in. After the missing fields are filled in, the integrity score is recalculated.

[0021] The data accuracy dynamic verification unit is used to randomly select verification samples from each batch of data, compare the samples with the benchmark database data, count the number of successfully matched records, calculate the matching accuracy rate, and simultaneously calculate the relative error of each sample for the numerical fields in the samples, and then calculate the average to obtain the numerical error rate. When the matching accuracy rate is less than the accuracy rate threshold or the numerical error rate is greater than or equal to the error rate threshold, the data verification fails, the data is marked as suspicious data, and pushed to the data provider for review. After the review confirms the error, it is corrected or deleted.

[0022] The trust assessment module includes a data provider trust assessment unit and a data accessor credibility rating unit.

[0023] The data provider trust assessment unit is used to calculate a trust score based on the historical data quality score, update timeliness score, and compliance score of the comprehensive data provider.

[0024] Furthermore, based on the historical data quality score, update timeliness score, and compliance of the comprehensive data provider, the steps for calculating the trust score include:

[0025] P1: The historical data quality score is calculated by weighting the data integrity score and accuracy score of the data provider for the past 6 months. ;

[0026] P2: Based on the government data update standards, calculate the ratio of the actual update delay days to the standard cycle to obtain the update timeliness score. ;

[0027] P3: Calculate the number of violations by the data provider over the past 6 months to obtain a compliance score. ;

[0028] P4: The trust score is calculated based on historical data quality score, update timeliness score, and compliance score. The specific calculation formula is as follows:

[0029] ;

[0030] Get a trust score ,in, All are weighting coefficients;

[0031] P5: When the trust score is greater than or equal to the trust threshold, it is determined to be a high-trust data provider; when the trust score is less than the trust threshold, it is determined to be a low-trust data provider; high-trust data providers simplify data review, while low-trust data providers conduct multiple rounds of verification on the provided data;

[0032] The data access party credibility rating unit calculates a credibility score based on the access party's historical behavior score, qualification level score, and usage compliance score. The specific calculation formula is as follows:

[0033] ;

[0034] Receive credibility score ,in, , and Let be the weighting coefficient, and satisfy . , Score historical behavior. To score for qualification level, Score for compliance with usage requirements;

[0035] The dynamic access control module includes a dynamic access permission determination unit and a real-time access behavior auditing unit.

[0036] The access permission dynamic determination unit is used to determine the access scope based on the access party's credibility score; when the credibility score is greater than or equal to the credibility threshold, the access permission to access advanced sensitive data is obtained; when the credibility score is lower than the credibility threshold, only public data can be accessed.

[0037] The real-time access behavior auditing unit is used to monitor the download volume, access frequency and operation type of the accessing party, identify unauthorized operations and issue alarms; when an alarm is triggered, the accessing party's permissions are immediately suspended and the alarm information is sent to the data management module. After manual review, the permissions are restored or permanently restricted.

[0038] The information sharing terminal includes a privacy computing module, a data fusion and sharing module, and a data management module;

[0039] The privacy computing module includes a federated learning parameter encryption unit and a differential privacy noise injection unit;

[0040] The federated learning parameter encryption unit uses homomorphic encryption technology to encrypt and transmit federated learning model parameters, thereby achieving data immobilization and model dynamics.

[0041] Furthermore, the steps for encrypting and aggregating federated learning model parameters based on homomorphic encryption technology to achieve data immobility and model mobility include:

[0042] O1: Establish a cross-departmental federated learning cluster, where data computing participants use local data to train models and obtain local model parameters;

[0043] O2: Data computation participants encrypt their local model parameters using a homomorphic encryption algorithm and upload the encrypted local model parameters to the coordinator to avoid parameter leakage during transmission;

[0044] O3: The coordinator calculates the weights based on the local model parameter data volume of the participating parties, and substitutes them into the formula to calculate the global model aggregation parameters. The specific calculation formula is as follows:

[0045] ;

[0046] Obtain the combined parameters of the global clustering model ,in, For the first The amount of data used to calculate the local model parameters of each participant. The total amount of data for all participants in the computation. For the first The encrypted local model parameters of each computation participant; the coordinator uses its private key to decrypt the global aggregated model parameters to obtain the global model parameters;

[0047] 04: Distribute the global model parameters to each data computation participant, update the local model, and repeat steps 01 to 03 until the global model parameters reach the confidence level;

[0048] The differential privacy noise injection unit is used to inject Laplace noise into the shared government statistical data to ensure that deleting or adding a record does not significantly change the statistical results. By injecting noise to obscure individual information, it ensures that attackers cannot infer individual data from the statistical results, while ensuring the overall availability of the statistical data.

[0049] The data fusion and sharing module includes a multi-source data association and fusion unit and a sharing permission dynamic allocation unit;

[0050] The multi-source data association and fusion unit calculates the correlation degree based on the key fields of government data. For data of the same subject, it merges them according to field priority to form a unified data subject after fusion. At the same time, it randomly selects 10% of the fused data and compares it with the original data of each department to verify the accuracy of the association. When the accuracy rate is lower than 95%, the weight of the associated fields is readjusted and fused again.

[0051] The shared permission dynamic allocation unit dynamically allocates data sharing permissions based on data sensitivity level, access party credibility score and business requirement level, realizing a dynamic allocation scheme of sharing on demand and minimum permissions.

[0052] The data management module is used to store system access data, and at the same time, it sends alarm information and data to be deleted to the administrator via email, and sends SMS to the administrator's terminal to remind the administrator to check the email content.

[0053] The technical effects and advantages of the information management system for a government information system based on data sharing and exchange according to the present invention are as follows:

[0054] 1. This invention addresses issues such as missing, redundant, and formatted data in government data by setting up a data processing terminal. It eliminates data noise through standardized processing and tracks the quality status of data in real time throughout the entire process of collection, transmission, and storage, quantitatively evaluating key indicators such as data integrity and accuracy.

[0055] 2. This invention, by setting up an information supervision terminal to quantify the credibility of data providers and users, accurately identifies the risk levels of different entities, provides a reliable basis for data sharing authorization, reduces security risks caused by low-credibility entities, and simultaneously adjusts the scope of access permissions and operation types in real time to achieve on-demand authorization. Furthermore, it promptly captures abnormal operations through behavior auditing, preventing the unauthorized acquisition or misuse of sensitive data while ensuring the efficient conduct of legitimate access, complying with data security regulations, and providing crucial support for the secure sharing of government data.

[0056] 3. This invention enables cross-departmental collaborative data computing without disclosing original government data by setting up an information sharing terminal, effectively avoiding the risk of privacy leakage, meeting the relevant data security regulations, providing technical support for multi-departmental joint efforts to optimize public services and analyze policy effects, solving the data collaboration problem caused by privacy concerns between departments, breaking down government data silos, achieving precise correlation and integration of multi-source data, forming a unified data subject, and improving the overall usability of data. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of the overall architecture of an information management system for a government information system based on data sharing and exchange, according to the present invention.

[0058] Figure 2 This is a schematic diagram of the data processing terminal architecture of an information management system for a government information system based on data sharing and exchange, according to the present invention.

[0059] Figure 3 This is a schematic diagram of the information supervision terminal architecture of an information management system for a government information system based on data sharing and exchange, according to the present invention.

[0060] Figure 4 This is a schematic diagram of the information sharing terminal architecture of an information management system for a government information system based on data sharing and exchange, according to the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” used in the embodiments of this invention are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0063] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0064] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0065] In practice, the server-side equipment deployed in an information management system for a data-sharing and exchange-based e-government information system may consist of one or more devices. This information management system can be implemented as a business instance, a virtual machine, or hardware devices. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node, providing a data-sharing and exchange-based e-government information management system to various user terminals. Alternatively, it can be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage various user terminals. Or, it can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide a data-sharing and exchange-based e-government information management system to various user terminals.

[0066] In terms of implementation, the information management system and user terminal of a government information system based on data sharing and exchange are mutually compatible. That is, if the information management system of a government information system based on data sharing and exchange is implemented as an application installed on a cloud service platform, then the user terminal is implemented as a client that establishes a communication connection with the application; or if the information management system of a government information system based on data sharing and exchange is implemented as a website, then the user terminal is implemented as a webpage; or if the information management system of a government information system based on data sharing and exchange is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0067] like Figure 1 The diagram shown is a system architecture diagram of an information management system for a government information system based on data sharing and exchange, provided by an embodiment of the present invention.

[0068] The information management system for a government information system based on data sharing and exchange, as described in this invention, can be located on a cloud server. In terms of implementation, it can function as one or more service devices, or as an application installed in the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed as a website. Depending on the functions implemented, the information management system for a government information system based on data sharing and exchange may include a data cleaning module, a data quality monitoring module, a trust assessment module, a dynamic access control module, a privacy computing module, and a data fusion and sharing module. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0069] In this embodiment of the invention, in an information management system for a government information system based on data sharing and exchange, each of the above-mentioned modules can be implemented independently and can call other modules. Here, "calling" can be understood as a module connecting to multiple modules of another type and providing corresponding services to those connected modules. For example, a sharing and evaluation module can call the same information collection module to obtain information collected by that module. Based on the above characteristics, in the information management system for a government information system based on data sharing and exchange provided by this embodiment of the invention, without modifying the program code, the applicable scope of the information management system architecture for a government information system based on data sharing and exchange can be adjusted by adding modules and directly calling them, achieving cluster-based horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the information management system for a government information system based on data sharing and exchange. In practical applications, the above-mentioned modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in a cloud server.

[0070] Example 1

[0071] Please see Figures 1 to 2 As shown in this embodiment, the information management system of a government information system based on data sharing and exchange includes: a data processing terminal, an information supervision terminal, and an information sharing terminal;

[0072] The data processing module, based on data cleaning, intelligently repairs missing values ​​and accurately removes redundant data, solving the problem of chaotic government data, achieving standardized processing, and providing a foundation for sharing; the data quality monitoring module monitors data integrity and accuracy in real time, ensuring the reliability of shared data and supporting precise and efficient government services.

[0073] The information supervision module, based on the trust assessment module, quantifies the quality, timeliness, and compliance of data providers, as well as the behavior, qualifications, and purpose of accessing parties, and grades the level of trust to provide a basis for data sharing authorization. The dynamic access control module calculates risks and adjusts permissions in real time based on data sensitivity and accessing party credibility, and audits behavior to prevent unauthorized access. The two work together to ensure the secure and controllable sharing of government information data.

[0074] The information sharing module, based on the privacy computing module, uses federated learning to encrypt and transmit model parameters and injects noise using differential privacy, enabling cross-departmental collaborative data computing without leaking original information and protecting individual privacy; the data fusion and sharing module associates multi-source data and dynamically allocates permissions to achieve precise matching and on-demand sharing, supporting the efficient operation of government scenarios;

[0075] The data processing module includes a data cleaning module and a data quality monitoring module;

[0076] The data cleaning module includes a missing value intelligent repair unit and a redundant data removal unit;

[0077] The intelligent missing value repair unit prioritizes government information data based on its importance. It then filters samples from the same batch of data that do not contain missing values ​​for the target missing field, excluding obvious outliers. For the filtered non-missing samples, it assigns corresponding weights based on the priority of their respective fields. The non-missing sample values ​​and their corresponding weights are then substituted into a formula to calculate the repair value. The specific calculation formula is as follows:

[0078] ;

[0079] Get repair value ,in, Let i be the field weight of the i-th non-missing sample. Let i be the target field value of the i-th non-missing sample. The total number of valid non-missing samples is used to calculate the repair value. The purpose of this calculation is to distinguish the importance of government affairs fields by weight, ensure the repair accuracy of core fields, and avoid cross-departmental data matching failures due to repair errors.

[0080] The redundant data removal unit selects key fields based on the data type of government affairs, and divides the key fields into text-based and numerical-based types. The similarity of text-based fields is calculated using word vector cosine similarity, while the similarity of numerical fields is calculated using relative error. At the same time, the similarity of each field is weighted and summed with its corresponding weight to calculate the comprehensive similarity. When the comprehensive similarity is greater than or equal to the redundancy threshold, it is determined to be redundant data. The most complete segment of the redundant data is retained, and the remaining redundant data is marked as data to be deleted and sent to the data management module for manual review and deletion.

[0081] It should be explained that the redundant value is manually set and entered; the purpose of weighted summation is to avoid misjudgment of a single field by weighting multiple fields, accurately clean up redundant data, and improve the accuracy and transmission efficiency of government data sharing.

[0082] The data quality monitoring module includes a data integrity monitoring unit and a data accuracy dynamic verification unit;

[0083] The data integrity monitoring unit calculates the field missing rate and record integrity score based on government data standards to monitor whether the data meets the basic requirements for sharing.

[0084] Furthermore, based on government data standards, the steps for calculating field missing rates and record integrity scores to monitor whether data meets basic sharing requirements include:

[0085] Step 1: Based on government data standards, determine the required and optional fields for each type of data; for each batch of data access, count the number of missing records for each field and calculate the field missing rate;

[0086] Step Two: Assign high weight to required fields and low weight to optional fields to ensure that the absence of core fields has a greater impact on completeness. Calculate the completeness score by substituting the values ​​into the formula. The specific calculation formula is as follows:

[0087] ;

[0088] Obtain integrity score ,in, This is the total number of fields required by the government data standard. Let k be the missing rate of the k-th field. The weight of the k-th field is used to calculate the integrity score to quantitatively assess data integrity, avoid the inability of shared data to be used due to the missing key fields, and balance monitoring accuracy and collection efficiency through weight.

[0089] Step 3: Based on the integrity score calculated in Step 2, a judgment is made. When the integrity score is less than the score threshold, a repair instruction is pushed to the data provider, specifying the missing fields and the time limit for filling them in. After the missing fields are filled in, the integrity score is recalculated.

[0090] It should be explained that the scoring threshold was set and entered manually; the repair instructions included re-collecting a new set of government information data.

[0091] The dynamic data accuracy verification unit is used to randomly select verification samples from each batch of data, compare the samples with the benchmark database, count the number of successfully matched records, calculate the matching accuracy rate, and simultaneously calculate the relative error of each sample for the numerical fields in the samples, and then calculate the average to obtain the numerical error rate. When the matching accuracy rate is less than the accuracy rate threshold or the numerical error rate is greater than or equal to the error rate threshold, the data verification fails, the data is marked as suspicious data, and pushed to the data provider for review. After the review confirms the error, it is corrected or deleted.

[0092] It should be explained that the sampling condition is to sample 10% of the total data in each batch; the accuracy threshold and error rate threshold are both manually set and entered.

[0093] Example 2

[0094] Please refer to Figure 3 As shown: Based on Embodiment 1, the information supervision terminal includes a trust assessment module and a dynamic access control module;

[0095] The trust assessment module includes a data provider trust assessment unit and a data accessor credibility rating unit;

[0096] The data provider trust assessment unit is used to calculate a trust score based on the historical data quality score, update timeliness score, and compliance score of the comprehensive data provider.

[0097] Furthermore, based on the historical data quality score, update timeliness score, and compliance of the comprehensive data provider, the steps for calculating the trust score include:

[0098] P1: The historical data quality score is calculated by weighting the data integrity score and accuracy score of the data provider for the past 6 months. ;

[0099] P2: Based on the government data update standards, calculate the ratio of the actual update delay days to the standard cycle to obtain the update timeliness score. ;

[0100] P3: Calculate the number of violations by the data provider over the past 6 months to obtain a compliance score. ;

[0101] P4: The trust score is calculated based on historical data quality score, update timeliness score, and compliance score. The specific calculation formula is as follows:

[0102] ;

[0103] Get a trust score ,in, All are weighting coefficients;

[0104] P5: When the trust score is greater than or equal to the trust threshold, it is determined to be a high-trust data provider; when the trust score is less than the trust threshold, it is determined to be a low-trust data provider; high-trust data providers simplify data review, while low-trust data providers conduct multiple rounds of verification on the provided data;

[0105] The data access party credibility rating unit calculates a credibility score based on the access party's historical behavior score, qualification level score, and usage compliance score. The specific calculation formula is as follows:

[0106] ;

[0107] Receive credibility score ,in, Score historical behavior. To score for qualification level, Score for compliance with usage requirements;

[0108] The dynamic access control module includes a dynamic access permission determination unit and a real-time access behavior auditing unit;

[0109] The dynamic access control unit is used to determine the access scope based on the accessor's credibility score; when the credibility score is greater than or equal to the credibility threshold, access to high-level sensitive data is granted; when the credibility score is lower than the credibility threshold, access to public data is only allowed.

[0110] It should be noted that the credibility threshold is manually set and entered.

[0111] The real-time access behavior auditing unit is used to monitor the download volume, access frequency and operation type of the accessing party, identify unauthorized operations and issue alarms; when an alarm is triggered, the accessing party's permissions are immediately suspended and the alarm information is sent to the data management module. After manual review, the permissions are restored or permanently restricted.

[0112] Example 3

[0113] Please refer to Figure 4 As shown: Based on Embodiment 1, the information sharing terminal includes a privacy computing module, a data fusion and sharing module, and a data management module;

[0114] The privacy computing module includes a federated learning parameter encryption unit and a differential privacy noise injection unit;

[0115] The federated learning parameter encryption unit uses homomorphic encryption technology to encrypt and transmit and aggregate federated learning model parameters, achieving data immobilization and model dynamics.

[0116] Furthermore, the steps for encrypting and aggregating federated learning model parameters based on homomorphic encryption technology to achieve data immobility and model mobility include:

[0117] O1: Establish a cross-departmental federated learning cluster, where data computing participants use local data to train models and obtain local model parameters;

[0118] O2: Data computation participants encrypt their local model parameters using a homomorphic encryption algorithm and upload the encrypted local model parameters to the coordinator to avoid parameter leakage during transmission;

[0119] O3: The coordinator calculates the weights based on the local model parameter data volume of the participating parties, and substitutes them into the formula to calculate the global model aggregation parameters. The specific calculation formula is as follows:

[0120] ;

[0121] Obtain the combined parameters of the global clustering model ,in, For the first The amount of data used to calculate the local model parameters of each participant. The total amount of data for all participants in the computation. For the first The encrypted local model parameters of each computation participant; the coordinator uses its private key to decrypt the global aggregated model parameters to obtain the global model parameters;

[0122] 04: Distribute the global model parameters to each data computation participant, update the local model, and repeat steps 01 to 03 until the global model parameters reach the confidence level;

[0123] The differential privacy noise injection unit is used to inject Laplace noise into shared government statistics data to ensure that deleting or adding a record does not significantly change the statistical results. By injecting noise to obscure individual information, it ensures that attackers cannot infer individual data from the statistical results, while ensuring the overall availability of the statistical data.

[0124] The data fusion and sharing module includes a multi-source data association and fusion unit and a dynamic allocation unit for sharing permissions;

[0125] The multi-source data association and fusion unit calculates the correlation degree based on the key fields of government data. For data of the same subject, it merges them according to field priority to form a unified data subject after fusion. At the same time, 10% of the fused data is randomly selected and compared with the original data of each department to verify the accuracy of the association. When the accuracy rate is lower than 95%, the weight of the associated fields is readjusted and fused again.

[0126] The shared permission dynamic allocation unit dynamically allocates data sharing permissions based on data sensitivity level, access party trust score and business requirement level, realizing a dynamic allocation scheme of sharing on demand and minimum permissions;

[0127] The data management module is used to store system access data, and at the same time, it sends alarm information and data to be deleted to administrators via email, and sends SMS messages to administrators' terminals to remind them to check the email content;

[0128] This embodiment effectively addresses the core pain points of data silos, inconsistent quality, and privacy leaks in government data sharing, demonstrating significant value across multiple dimensions. At the data governance level, the data cleaning module repairs missing data and removes redundancy, while the data quality monitoring module provides real-time monitoring and anomaly tracing, significantly improving the standardization and reliability of government data and laying a high-quality foundation for cross-departmental data sharing. At the security level, the trust assessment module quantifies the risks for both data providers and users, and the dynamic access control module adjusts permissions in real-time, while the privacy computing module provides privacy protection, constructing a multi-layered security system. This system precisely controls the scope of data access, effectively preventing unauthorized data use and privacy leaks, and ensuring security throughout the entire data sharing process. At the sharing efficiency level, the data fusion and sharing module integrates multi-source data and allocates permissions as needed, promoting efficient cross-departmental data collaboration, significantly shortening the data retrieval cycle, and providing reliable technical support for optimizing the government data management system and building a digital government.

[0129] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the present invention.

Claims

1. An information management system for a government information system based on data sharing exchange, characterized by, The system comprises a data processing end, an information supervision end and an information sharing end; The data processing end solves the problem of disordered government data by intelligently repairing missing values and accurately removing redundant data based on a data cleaning module, realizes standardized processing and provides a basis for sharing; the data quality monitoring module monitors the data integrity and accuracy in real time to ensure the reliability of shared data and support accurate and efficient government services; The information supervision end provides a basis for data sharing authorization by quantifying the quality, timeliness, compliance of data providers and the behavior, qualifications and purposes of access parties, and grading the trustworthiness based on a trust evaluation module; the dynamic access control module calculates the risk in real time based on the data sensitivity and the credibility of the access party, adjusts the permissions, audits the behavior to prevent unauthorized access, and cooperates to ensure the safe and controllable sharing of government information data; The information sharing end protects individual privacy by implementing collaborative computing without leaking original information through federated learning and encrypted transmission of model parameters and differential privacy noise based on a privacy computing module; the data fusion and sharing module correlates multi-source data and dynamically allocates permissions to realize accurate matching and on-demand sharing, and supports efficient operation of government scenarios.

2. The information management system of the government information system based on data sharing exchange according to claim 1, wherein, The data processing end comprises a data cleaning module and a data quality monitoring module; The data cleaning module comprises an intelligent missing value repair unit and a redundant data removal unit; The intelligent missing value repair unit divides information priority based on the importance of government information data, filters non-missing samples of the missing field in the same batch of data for the target missing field, and excludes obvious outliers; The non-missing samples are assigned corresponding weights according to their field priority, and the non-missing sample values and corresponding weights are substituted into the formula to calculate the repair value, and the specific calculation formula is: ； The repair value is obtained wherein, is the field weight of the ith non-missing sample, is the target field value of the ith non-missing sample, is the total number of valid non-missing samples, and the function of calculating the repair value is to distinguish the importance of the government field through the weight, ensure the repair accuracy of the core field, and avoid cross-department data matching failure caused by repair errors. The redundant data removal unit selects key fields based on the type of government data, and divides the key fields into text type and numerical type; the similarity of text type fields is calculated by word vector cosine, and the similarity of numerical type fields is calculated by relative error; the comprehensive similarity is calculated by weighting the single field similarity and the corresponding weight, and when the comprehensive similarity is greater than or equal to the redundancy threshold, it is determined as redundant data; the most complete part of the redundant data is retained, and the remaining redundant data is marked as to-be-deleted data and sent to the data management module for deletion after manual review.

3. The information management system of the government affair information system based on data sharing exchange according to claim 2, characterized in that, The data quality monitoring module comprises a data integrity monitoring unit and a data accuracy dynamic verification unit; The data integrity monitoring unit calculates the field missing rate and record integrity score based on government data standards to monitor whether the data meets the basic requirements for sharing; Further, the steps of calculating the field missing rate and record integrity score based on government data standards to monitor whether the data meets the basic requirements for sharing comprise: Step one: based on the government data standards, determine the mandatory fields and optional fields of each type of data; for each batch of access data, count the number of missing records for each field and calculate the field missing rate; Step two: assign high weight to mandatory fields and low weight to optional fields to ensure that the missing of core fields has a greater impact on integrity; calculate the integrity score by substituting the formula, and the specific calculation formula is: ； obtaining an integrity score wherein, is a total number of fields required by the government data standard, is a missing rate of the kth field, is a weight of the kth field, calculating the integrity score is used to quantify and assess data integrity, avoid sharing data from being unable to use due to missing of key fields, and balance monitoring precision and collection efficiency through the weight. Step three: based on the integrity score calculated in step two, when the integrity score is less than the score threshold, push repair instructions to the data provider, specify the missing field and the filling time limit, and recalculate the integrity score after filling; The data accuracy dynamic checking unit is used for randomly extracting checking samples from each batch of data, comparing the samples with the reference library data, counting the number of matching records, calculating the matching accuracy, and at the same time, for the numerical fields in the samples, calculating the relative error of each sample, and then averaging to obtain the numerical error rate; When the matching accuracy is less than the accuracy threshold or the numerical error rate is greater than or equal to the error rate threshold, the data checking fails, the data is marked as suspicious data, and is pushed to the data provider for review, and after the review confirms the error, it is modified or deleted.

4. The information management system of the government information system based on data sharing exchange according to claim 1, wherein, The information supervision end includes a trust evaluation module and a dynamic access control module; The trust evaluation module includes a data provider trust degree evaluation unit and a data access party credibility rating unit; The data provider trust degree evaluation unit is used for calculating the trust degree score based on the historical data quality score, the update timeliness score and the compliance score of the comprehensive data provider; Further, the step of calculating the trust degree score based on the historical data quality score, the update timeliness score and the compliance score of the comprehensive data provider includes: P1: Based on the historical 6-month data, the data integrity score of the provider is weighted with the accuracy rate to calculate the historical data quality score ; P2: Based on the government data update standard, calculate the ratio of the actual update delay days to the standard period to obtain the update timeliness score ; P3: Count the number of violations of the data provider in the last 6 months to get the compliance score ; P4: Based on the historical data quality score, the update timeliness score and the compliance score, the trust degree score is calculated, and the specific calculation formula is: ； obtaining a trust score wherein, are weight coefficients; P5: When the trust degree score is greater than or equal to the trust degree threshold, it is determined as a high-trust data provider; When the trust degree score is less than the trust degree threshold, it is determined as a low-trust data provider; The high-trust data provider simplifies data review, and the low-trust data provider performs multiple rounds of verification on the provided data; The data access party credibility rating unit calculates the credibility score based on the historical behavior score, the qualification level score and the use compliance score of the access party, and the specific calculation formula is: ； obtaining a trustworthiness score wherein, , and are weight coefficients, and satisfy , is a historical behavior score, is a qualification level score, is a use compliance score.

5. The information management system of the government information system based on data sharing exchange according to claim 4, characterized in that, The dynamic access control module includes an access permission dynamic determination unit and an access behavior real-time audit unit; The access permission dynamic determination unit is used for determining the access range based on the credibility score of the access party; When the credibility score is greater than or equal to the credibility threshold, the permission to access high-level sensitive data is obtained; When the credibility score is lower than the credibility threshold, only public data can be accessed; The access behavior real-time audit unit is used for monitoring the download volume, access frequency and operation type behavior of the access party, identifying unauthorized operations and alarming; When the alarm is triggered, the access party's permission is immediately suspended, and the alarm information is sent to the data management module, and after manual review, the permission is restored or permanently limited.

6. The information management system of the government information system based on data sharing exchange according to claim 1, wherein, The information sharing end includes a privacy calculation module, a data fusion sharing module and a data management module; The privacy calculation module includes a federated learning parameter encryption unit and a differential privacy noise injection unit; The federated learning parameter encryption unit encrypts the transmission and aggregation of federated learning model parameters based on homomorphic encryption technology, realizes data motionless model motion, and the specific steps include: Further, the step of encrypting the transmission and aggregation of federated learning model parameters based on homomorphic encryption technology to realize data motionless model motion includes: O1: Establish a cross-department federated learning cluster, and data computing participants train models using local data to obtain local model parameters; O2: Data computing participants encrypt local model parameters based on homomorphic encryption algorithms and upload encrypted local model parameters to the coordinator to avoid parameter leakage during transmission; O3: The coordinator calculates the weight based on the data volume of the local model parameters of the data computing participants, and substitutes it into the formula to calculate the global model aggregation parameters, and the specific calculation formula is: ； obtain the global aggregated model parameter wherein, the total data amount of all computing participants, the data amount of the local model parameter of the computing participant, the total data amount of all computing participants, the encrypted local model parameter of the computing participant, the global model parameter obtained by the coordinator by decrypting the global aggregated model parameter using the private key. 04: Distribute the global model parameters to each data computing participant to update the local model, and repeat steps 01 to 03 until the global model parameters reach the confidence level; The differential privacy noise injection unit is used to inject Laplace noise into the shared government statistical data, ensuring that deleting or adding a record does not significantly change the statistical results, and by injecting noise to obscure individual information, it ensures that attackers cannot infer individual data from the statistical results, while ensuring the overall usability of the statistical data.

7. The information management system of the government information system based on data sharing exchange according to claim 6, characterized in that, The data fusion sharing module includes a multi-source data association fusion unit and a shared permission dynamic allocation unit; The multi-source data association fusion unit calculates the correlation degree based on the key fields of government data, merges the data of the same subject according to the field priority, forms the unified data subject after fusion, and randomly extracts 10% of the fused data for comparison with the original data of each department to verify the accuracy of the association. When the accuracy is less than 95%, the association field weight is adjusted and fused again. The shared permission dynamic allocation unit dynamically allocates data sharing permissions based on data sensitivity levels, access party credibility scores, and business demand levels, implementing a dynamic allocation scheme of on-demand sharing and minimum permissions.

8. The information management system of the government information system based on data sharing exchange according to claim 6, wherein, The data management module is used to store system access data, and sends alarm information and to-be-deleted data to the administrator through email and sends a short message to the administrator's terminal to remind the administrator to check the email content.

Images