Efficiency improvement method for government-enterprise digital management

Abstract

This invention discloses an efficiency improvement method for digital management of government and enterprises, relating to the field of big data analytics. The method includes: S1, establishing a business architecture interaction data synchronization network for various project types within the government and enterprise, marking the execution processes of transaction logs for each project type, generating a dynamic permission network for business architecture interaction for each project type, verifying the dependencies between each transaction execution process under the dynamic permissions of the business architecture, assigning priorities to transaction execution processes, determining the initial resource allocation for each project type, establishing a resource allocation benefit evaluation model for each project type, and generating the optimal resource allocation scheme for each project type. The beneficial effects of this invention are: significantly improving the systematicness and adaptability of digital management for government and enterprises, and providing a holistic solution that balances compliance and efficiency for complex business scenarios.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, specifically to a method for improving efficiency in digital management for government and enterprises. Background Technology

[0002] Currently, digital management in government and enterprises suffers from severe data silos, making it difficult to effectively integrate heterogeneous data from multiple sources; rigid access control mechanisms fail to dynamically adapt to business needs; resource allocation relies on manual experience and lacks scientific quantitative basis; linear and rigid transaction processing workflows fail to identify critical path priorities; and evaluation systems are one-dimensional, ignoring the multidimensional relationship between resource input and value output, leading to problems such as low management efficiency and unreasonable resource allocation. Summary of the Invention

[0003] To address the aforementioned technical issues and provide methods for improving efficiency in digital management for government and enterprises, this technical solution resolves the problems described above.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] Efficiency improvement methods for digital management in government and enterprises include:

[0006] S1. Obtain multi-source heterogeneous data of various project types of government and enterprises, establish a business architecture knowledge graph of various project types of government and enterprises, perform semantic association according to the business architecture of various project types of government and enterprises, and generate a business architecture interaction data synchronization network of various project types of government and enterprises.

[0007] S2. Based on the business architecture interaction data synchronization network of various government and enterprise project types, mark the execution process of transaction logs of various government and enterprise project types, establish a multi-level dynamic permission allocation model, and generate a dynamic permission network for business architecture interaction of various government and enterprise project types.

[0008] S3. Based on the dynamic permission network of the business architecture interaction of various government and enterprise project types, verify the dependency relationship between each transaction execution process under the dynamic permission of the business architecture, establish a transaction execution process priority identification mechanism, assign priority to the transaction execution process, and determine the initial resource allocation of the transaction execution process of various government and enterprise project types.

[0009] S4. Screen the correlation between the value and benefits of resource allocation in the execution process of various government and enterprise projects in history, establish a resource allocation benefit evaluation model for various government and enterprise projects, generate the optimal range of resource allocation-value benefit for the execution process of various government and enterprise projects, and revise the initial resource allocation for the execution process of various government and enterprise projects to obtain the optimal resource allocation scheme for each government and enterprise project type.

[0010] Preferably, step S1 specifically includes:

[0011] Based on multi-source heterogeneous data from various government and enterprise projects, the data is divided into structured, semi-structured, and unstructured data. A multi-head attention mechanism decoder is used to match the corresponding data types to generate a unified spatiotemporal reference data set for various government and enterprise project types.

[0012] Based on a unified spatiotemporal reference dataset for various government and enterprise project types, the data is substituted into a BiLSTM-CRF joint relation extraction model. The BiLSTM is used to label the contextual semantic information of each structural data point within the unified spatiotemporal reference dataset for each government and enterprise project type at a given time, which is then input into the CRF. The conditional probabilities of entity / dependency labels corresponding to the contextual semantic information of each structural data point at a given time are calculated, thus determining the data entity-relationship probability distribution for each government and enterprise project type. The method is as follows:

[0013] ;

[0014] in, This refers to the conditional probabilities of entity / dependency labels given unified spatiotemporal benchmark data for various types of government and enterprise projects. A binary variable for entity / dependency label, This provides the contextual semantic information of each structural data point in the unified spatiotemporal reference data for various government and enterprise project types at the t-th unit of time. This is a weighted matrix of entity / dependency labels. The entity / dependency label for the t-th unit of time is transferred to the t-th unit of time. Entity / dependency label offset per unit time, This refers to the sequence of all possible entity / dependency labels when providing unified spatiotemporal benchmark data for various government and enterprise project types.

[0015] Preferably, step S1 further includes:

[0016] Based on the data entity-relationship probability distribution of each type of government and enterprise project, and according to the business architecture level of each type of government and enterprise project, the data entity-relationship probability distribution of the corresponding business architecture level is substituted into the GNN graph neural network. The business architecture level is used as the global entity, the data entity of the corresponding type of government and enterprise project is used as the internal child node, and the data dependency relationship of the corresponding type of government and enterprise project is used as the edge weight to establish a data self-synchronization network of the business architecture of each type of government and enterprise project.

[0017] Based on Jaccard-EditDistance hybrid similarity, this paper uses the Jaccard similarity algorithm to quantify the text character similarity of child nodes within the data self-synchronization network of business architectures for various government and enterprise project types. It then uses EditDistance to quantify the text character operation change similarity of these child nodes, determining the hybrid similarity between child nodes across different business architectures. This establishes an interactive data synchronization network for the business architectures of various government and enterprise project types, as follows:

[0018] ;

[0019] in, Let be the mixed similarity between the k-th child node and the (k+1)-th child node within the data self-synchronization network of the business architecture of the i-th project type in government and enterprise. These are the weighting coefficients. The text character similarity of the k-th child node within the data self-synchronization network of the business architecture for the i-th project type in government and enterprise. The similarity of text character operation changes within the k-th child node of the data self-synchronization network for the business architecture of the i-th project type in government and enterprise.

[0020] Preferably, step S2 specifically includes:

[0021] The execution process of the transaction logs of various government and enterprise project types is marked as the key-value attribute feature of the internal nodes in the business architecture interaction data synchronization network of various government and enterprise project types, thereby obtaining the sequence of sub-node transaction processes in the business architecture interaction data synchronization network of various government and enterprise project types.

[0022] Based on the actual constraints of the business architecture of various government and enterprise projects, establish restrictions on transaction process state transitions.

[0023] Based on the LSTM (Long Short-Term Memory) network, a multi-level dynamic permission allocation model is established. According to the transaction process state transition constraints, the interaction path constraints between the sub-nodes of the dynamic permission network for the interaction of business architectures of various government and enterprise project types are assigned. The transaction process sequence of the sub-nodes in the data synchronization network for the interaction of business architectures of various government and enterprise project types is used as input, and the data synchronization constraint network for the interaction of business architectures of various government and enterprise project types is used as output.

[0024] Using NMTF nonnegative matrix trifactoring, we verify the strength of the dependency relationship between the transaction process pointing edges of internal child nodes in the network of business architecture interaction data synchronization constraints for various project types in government and enterprises.

[0025] By utilizing the CVAE conditional variational autoencoder, the transaction process context information of internal child nodes in the business architecture interaction data synchronization constraint network for various government and enterprise project types is verified. The dependency strength of the transaction process pointing edges of internal child nodes is used as the latent vector. Using the reconstruction error and KL divergence, a permission allocation loss function is constructed to minimize the reconstruction error of the permission allocation loss function, the error between the actual permission allocation and the difference between the generated distribution and the prior distribution of KL divergence. Dynamic permissions are assigned to the transaction process sequences of child nodes in the business architecture interaction data synchronization network for various government and enterprise project types, resulting in a dynamic permission network for business architecture interaction for various government and enterprise project types.

[0026] Preferably, step S3 specifically includes:

[0027] Based on the dynamic permission network of business architecture interaction for various government and enterprise project types, obtain the transaction process and its dynamic permission tag in the business architecture of various government and enterprise project types.

[0028] Based on recursive causal discovery, the causal relationship between the transaction process and dynamic permission tag of the internal child node in the dynamic permission network of the business architecture interaction of various project types of government and enterprise is calculated, the dependency path between the transaction processes of the internal child node is determined, and the dynamic permission dependency relationship network of the business architecture interaction of various project types of government and enterprise is obtained.

[0029] Based on the dynamic permission dependency network of business architecture interaction for various government and enterprise project types, the transaction process request open metadata of internal sub-nodes under the dynamic permissions of business architecture interaction is obtained and normalized to obtain the transaction process request feature data of internal sub-nodes.

[0030] Using the analytic hierarchy process (AHP), the strength of the dependency relationship between the transaction process pointing edges of the internal sub-nodes in the network is constrained according to the business architecture interaction data synchronization constraints of various government and enterprise project types, and the transaction process request feature data weights of the internal sub-nodes are assigned.

[0031] The priority of the transaction process of internal sub-nodes in the network is determined by weighting and fusing the transaction process request feature data weights of internal sub-nodes with the transaction process request feature data of internal sub-nodes for business architecture interaction data synchronization constraints of various government and enterprise project types.

[0032] Preferably, step S3 further includes:

[0033] Based on the total amount of allocable resources per unit time for the business architecture of various government and enterprise project types, resource pool constraints are established for each government and enterprise project type.

[0034] Based on the business architecture interaction data synchronization constraints of various government and enterprise project types, the transaction process priority vector of internal sub-nodes in the network is constructed.

[0035] Normalize the transaction process priority vectors of internal child nodes;

[0036] Determine the scope of resource requests for the transaction processes of internal sub-nodes in the network, which are subject to the business architecture interaction data synchronization constraints for various project types in government and enterprises.

[0037] The objective function is to maximize the allocation of resource pools for each type of government and enterprise project to meet the business architecture interaction data synchronization constraints of each type of government and enterprise project. The resource pool constraints of each type of government and enterprise project are used as the total constraint. The normalized value of the priority vector of the transaction process of the internal sub-nodes is used as the resource allocation decision variable to determine the initial resource allocation of the transaction execution process of each type of government and enterprise project.

[0038] Preferably, step S4 specifically includes:

[0039] Obtain historical data on the execution progress and resource allocation of various government and enterprise projects.

[0040] Leveraging Bayesian networks, we can verify the posterior probability of the value and benefit impact of the transaction process request feature data of internal sub-nodes under the resource allocation of the transaction execution process for various types of government and enterprise projects, and determine the correlation matrix of resource allocation and value and benefit impact factors for the transaction execution process of various types of government and enterprise projects.

[0041] Using principal component analysis (PCA), the correlation matrix of resource allocation and value return influencing factors in the execution process of various government and enterprise projects is reduced in dimensionality to obtain the dimensionality-reduced correlation matrix of resource allocation and value return influencing factors in the execution process of various government and enterprise projects.

[0042] Based on random forest, a resource allocation benefit evaluation model for various types of government and enterprise projects is established, with the dimensionality reduction matrix of the correlation between resource allocation and value benefit influencing factors in the transaction execution process as the root node, the resource allocation and value benefit influencing factors in the transaction execution process as the branch nodes, and the optimal range of resource allocation and value benefit in the transaction execution process as the leaf nodes.

[0043] The initial resources of the transaction execution process of each type of government and enterprise project are used as the input of the resource allocation benefit evaluation model of each type of government and enterprise project. The constraint-sensitive split quality function is used as the split criterion to generate the optimal resource allocation scheme of each type of government and enterprise project as the output.

[0044] Specifically, the constraint-sensitive splitting quality function is:

[0045] ;

[0046] in, Let the terms be binary variables: the term for maximizing the purity of profit and the term for satisfying resource constraints. This represents the value corresponding to the transaction process request characteristic data of the k-th internal child node under the resource allocation of the transaction execution process for the i-th project type in historical government and enterprise projects. The average return on the execution process of various types of government and enterprise projects throughout history. A dimension-reduced matrix showing the correlation between resource allocation and value-benefit influencing factors in the execution process of various government and enterprise projects. This refers to the resource request vectors for the execution processes of various project types in government and enterprise sectors. The total amount of resources in the government and enterprise project pool for each project type. Resource allocation constraint parameters, This represents the total number of internal child nodes.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] This invention proposes an efficiency improvement solution for digital management in government and enterprises. It achieves deep integration of multi-source data by constructing a business architecture knowledge graph, breaking down information silos; it enables intelligent real-time permission adjustment based on a dynamic permission network, improving security and compliance; it establishes a scientific resource allocation mechanism through transaction priority identification and causal analysis, optimizing key business processes; and finally, it combines a value-benefit assessment model to form a closed-loop optimization system, ensuring that resource allocation meets both efficiency requirements and maximizes business value. The entire solution achieves intelligent management across the entire chain, from data governance to decision optimization, significantly improving the systematicness and adaptability of digital management in government and enterprises, and providing a comprehensive solution that balances compliance and efficiency for complex business scenarios. Attached Figure Description

[0049] Figure 1 A flowchart illustrating methods for improving efficiency in government and enterprise digital management. Detailed Implementation

[0050] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0051] Reference Figure 1 As shown, the efficiency improvement methods for digital management in government and enterprises include:

[0052] S1. Obtain multi-source heterogeneous data of various project types of government and enterprises, establish a business architecture knowledge graph of various project types of government and enterprises, perform semantic association according to the business architecture of various project types of government and enterprises, and generate a business architecture interaction data synchronization network of various project types of government and enterprises.

[0053] Step S1 specifically includes:

[0054] Based on multi-source heterogeneous data from various government and enterprise project types, the data is categorized into structured, semi-structured, and unstructured data. A multi-head attention mechanism decoder is used to match the corresponding data types, generating a unified spatiotemporal benchmark data set for each government and enterprise project type. The structured, semi-structured, and unstructured data include: structured data (database tables, Excel files) for each government and enterprise project type; structured data (JSON / XML logs) for each government and enterprise project type; and unstructured data (contract texts, meeting minutes) for each government and enterprise project type.

[0055] Based on a unified spatiotemporal reference dataset for various government and enterprise project types, the data is substituted into a BiLSTM-CRF joint relation extraction model. The BiLSTM is used to label the contextual semantic information of each structural data point within the unified spatiotemporal reference dataset for each government and enterprise project type at a given time, which is then input into the CRF. The conditional probabilities of entity / dependency labels corresponding to the contextual semantic information of each structural data point at a given time are calculated, thus determining the data entity-relationship probability distribution for each government and enterprise project type. The method is as follows:

[0056] ;

[0057] in, This refers to the conditional probabilities of entity / dependency labels given unified spatiotemporal benchmark data for various types of government and enterprise projects. A binary variable for entity / dependency label, This provides the contextual semantic information of each structural data point in the unified spatiotemporal reference data for various government and enterprise project types at the t-th unit of time. This is a weighted matrix of entity / dependency labels. The entity / dependency label for the t-th unit of time is transferred to the t-th unit of time. Entity / dependency label offset per unit time, This refers to the sequence of all possible entity / dependency labels when providing unified spatiotemporal benchmark data for various government and enterprise project types.

[0058] Step S1 also includes:

[0059] Based on the data entity-relationship probability distribution of each type of government and enterprise project, and according to the business architecture level of each type of government and enterprise project, the data entity-relationship probability distribution of the corresponding business architecture level is substituted into the GNN graph neural network. The business architecture level is used as the global entity, the data entity of the corresponding type of government and enterprise project is used as the internal child node, and the data dependency relationship of the corresponding type of government and enterprise project is used as the edge weight to establish a data self-synchronization network of the business architecture of each type of government and enterprise project.

[0060] Based on Jaccard-EditDistance hybrid similarity, this paper uses the Jaccard similarity algorithm to quantify the text character similarity of child nodes within the data self-synchronization network of business architectures for various government and enterprise project types. It then uses EditDistance to quantify the text character operation change similarity of these child nodes, determining the hybrid similarity between child nodes across different business architectures. This establishes an interactive data synchronization network for the business architectures of various government and enterprise project types, as follows:

[0061] ;

[0062] in, Let be the mixed similarity between the k-th child node and the (k+1)-th child node within the data self-synchronization network of the business architecture of the i-th project type in government and enterprise. These are the weighting coefficients. The text character similarity of the k-th child node within the data self-synchronization network of the business architecture for the i-th project type in government and enterprise. The similarity of text character operation changes within the k-th child node of the data self-synchronization network for the business architecture of the i-th project type in government and enterprise.

[0063] When using it, please refer to the steps outlined above:

[0064] As a further development, this solution aligns heterogeneous data from multiple sources within government and enterprises through a multi-head attention mechanism, utilizes a BiLSTM-CRF model to extract entity relationships with high precision and calculate probability distributions, constructs a business architecture knowledge graph based on a GNN graph neural network to achieve dynamic synchronization, and combines Jaccard-EditDistance hybrid similarity to achieve cross-system data consistency verification. Its core strength lies in achieving automated fusion of structured / unstructured data, reducing manual intervention, improving entity relationship recognition accuracy compared to traditional methods, optimizing the dynamic response speed of the business architecture from hours to minutes, and reducing cross-system field matching error rates.

[0065] For example:

[0066] Scenario: Digital management of "infrastructure project approval" in a municipal finance bureau

[0067] Input data: Structured: Project budget table in Oracle database; Semi-structured: JSON logs from the bidding system; Unstructured: Scanned PDF copies of contracts.

[0068] S1 execution flow:

[0069] Data unification: The multi-head attention model maps “budget amount” (Excel), “bid_amount” (JSON), and “total contract price” (PDF) to a unified field “project_budget”.

[0070] Knowledge extraction: BiLSTM-CRF extracts triples <Municipal Finance Bureau, Approval, Infrastructure Project A> from the contract text with a confidence level of 0.92.

[0071] Graph Construction: GNN associates the "infrastructure approval" business architecture with entities such as "Municipal Finance Bureau", "Project A", and "Budget", and the edge weights reflect the strength of the approval process dependency.

[0072] Dynamic synchronization: The similarity between "bid_amount=120 million" in the JSON log and "total price of one hundred and twenty million yuan" in the contract text is 0.98, and data correction is automatically triggered.

[0073] Output: Generates a dynamically updated "Infrastructure Project Approval" business architecture synchronization network, supporting real-time data collaboration across departments.

[0074] S2. Based on the business architecture interaction data synchronization network of various government and enterprise project types, mark the execution process of transaction logs of various government and enterprise project types, establish a multi-level dynamic permission allocation model, and generate a dynamic permission network for business architecture interaction of various government and enterprise project types.

[0075] Step S2 specifically includes:

[0076] The execution process of the transaction logs of various government and enterprise project types is marked as the key-value attribute feature of the internal nodes in the business architecture interaction data synchronization network of various government and enterprise project types, thereby obtaining the sequence of sub-node transaction processes in the business architecture interaction data synchronization network of various government and enterprise project types.

[0077] Based on the actual constraints of the business architecture of various government and enterprise projects, establish restrictions on transaction process state transitions.

[0078] As a further example, a business architecture constraint is as follows: the procurement department can only proceed with procurement transactions after financial approval is completed, and the corresponding transaction process state transition restriction condition is: financial approval completion triggers procurement execution;

[0079] Based on the LSTM (Long Short-Term Memory) network, a multi-level dynamic permission allocation model is established. According to the transaction process state transition constraints, the interaction path constraints between the sub-nodes of the dynamic permission network for the interaction of business architectures of various government and enterprise project types are assigned. The transaction process sequence of the sub-nodes in the data synchronization network for the interaction of business architectures of various government and enterprise project types is used as input, and the data synchronization constraint network for the interaction of business architectures of various government and enterprise project types is used as output.

[0080] Using NMTF nonnegative matrix trifactoring, we verify the strength of the dependency relationship between the transaction process pointing edges of internal child nodes in the network of business architecture interaction data synchronization constraints for various project types in government and enterprises.

[0081] By utilizing the CVAE conditional variational autoencoder, the transaction process context information of internal child nodes in the business architecture interaction data synchronization constraint network for various government and enterprise project types is verified. The dependency strength of the transaction process pointing edges of internal child nodes is used as the latent vector. Using the reconstruction error and KL divergence, a permission allocation loss function is constructed to minimize the reconstruction error of the permission allocation loss function, the error between the actual permission allocation and the difference between the generated distribution and the prior distribution of KL divergence. Dynamic permissions are assigned to the transaction process sequences of child nodes in the business architecture interaction data synchronization network for various government and enterprise project types, resulting in a dynamic permission network for business architecture interaction for various government and enterprise project types.

[0082] When using it, refer to the steps outlined above.

[0083] As a further development, this solution marks transaction logs as key-value attributes of the business architecture knowledge graph, utilizes LSTM state machines to model constrained process state transitions, combines Non-negative Matrix Trifactoring (NMTF) to mine internal node-permission relationships in the network architecture, and employs Conditional Variational Autoencoder (CVAE) to generate context-aware dynamic permission allocation, ultimately constructing a dynamic permission network strongly consistent with business rules. Its core strength lies in achieving dual spatiotemporal dynamism of permissions: the time dimension ensures operational timing compliance through LSTM memory units, the spatial dimension optimizes cross-departmental permission granularity through NMTF, and CVAE quantifies business risks through latent vector quantification to achieve adaptive security policies. Compared to the traditional RBAC model, this solution reduces permission redundancy and increases the interception rate of illegal operations, making it particularly suitable for government and enterprise digital management scenarios requiring real-time permission adjustments and strict compliance requirements, such as government procurement and cross-departmental collaborative projects. It significantly reduces manual policy maintenance costs while improving security.

[0084] For example:

[0085] Scenario and business architecture: Financial approval → Procurement bidding → Contract signing → Performance acceptance

[0086] Example of a transaction log: json{"Process ID":"P-2025-001","Department":"Finance Department","Operation":"Budget Approval","Status":"Completed","Time":"2025-05-10T14:30:00"};

[0087] Permission network generation process:

[0088] 1. Feature tagging: Map the logs to the dynamic attributes of the knowledge graph node "Financial Approval": {"Status": "Completed", "Operator": "Li XX"};

[0089] 2. State transition constraints: In the rule matrix, {{Financial Approval}, {Procurement Bidding}} = 1, and other procurement-related transitions are initially set to 0;

[0090] 3. Dynamic Permission Assignment: NMTF analysis revealed that the "Bidding Specialist" role needs to be associated with both the Finance Department (for viewing budgets) and the Procurement Office (for issuing tenders). CVAE generates the following permissions for this role: {"Actionable Processes":["Procurement and Bidding"], "Data Access Scope":["Budget Amount"]}.

[0091] Conflict detection: When the same user applies for both "Tender Execution" and "Budget Modification" permissions at the same time, manual review is triggered because the KL divergence exceeds the threshold.

[0092] S3. Based on the dynamic permission network of the business architecture interaction of various government and enterprise project types, verify the dependency relationship between each transaction execution process under the dynamic permission of the business architecture, establish a transaction execution process priority identification mechanism, assign priority to the transaction execution process, and determine the initial resource allocation of the transaction execution process of various government and enterprise project types.

[0093] Step S3 specifically includes:

[0094] Based on the dynamic permission network of business architecture interaction for various government and enterprise project types, obtain the transaction process and its dynamic permission tag in the business architecture of various government and enterprise project types.

[0095] Based on recursive causal discovery, the causal relationship between the transaction process and dynamic permission tag of the internal child node in the dynamic permission network of the business architecture interaction of various project types of government and enterprise is calculated, the dependency path between the transaction processes of the internal child node is determined, and the dynamic permission dependency relationship network of the business architecture interaction of various project types of government and enterprise is obtained.

[0096] Based on the dynamic permission dependency network of business architecture interaction for various government and enterprise project types, the transaction process request open metadata of internal sub-nodes under the dynamic permissions of business architecture interaction is obtained and normalized to obtain the transaction process request feature data of internal sub-nodes.

[0097] Using the analytic hierarchy process (AHP), the strength of the dependency relationship between the transaction process pointing edges of the internal sub-nodes in the network is constrained according to the business architecture interaction data synchronization constraints of various government and enterprise project types, and the transaction process request feature data weights of the internal sub-nodes are assigned.

[0098] The priority of the transaction process of internal sub-nodes in the network is determined by weighting and fusing the transaction process request feature data weights of internal sub-nodes with the transaction process request feature data of internal sub-nodes for business architecture interaction data synchronization constraints of various government and enterprise project types.

[0099] Step S3 also includes:

[0100] Based on the total amount of allocable resources per unit time for the business architecture of various government and enterprise project types, resource pool constraints are established for each government and enterprise project type.

[0101] Based on the business architecture interaction data synchronization constraints of various government and enterprise project types, the transaction process priority vector of internal sub-nodes in the network is constructed.

[0102] Normalize the transaction process priority vectors of internal child nodes;

[0103] Determine the scope of resource requests for the transaction processes of internal sub-nodes in the network, which are subject to the business architecture interaction data synchronization constraints for various project types in government and enterprises.

[0104] The objective function is to maximize the allocation of resource pools for each type of government and enterprise project to meet the business architecture interaction data synchronization constraints of each type of government and enterprise project. The resource pool constraints of each type of government and enterprise project are used as the total constraint. The normalized value of the priority vector of the transaction process of the internal sub-nodes is used as the resource allocation decision variable to determine the initial resource allocation of the transaction execution process of each type of government and enterprise project.

[0105] When using it, refer to the steps outlined above.

[0106] As a further development, a dynamic permission network and causal discovery are used to construct transaction process dependencies. The Analytic Hierarchy Process (AHP) is combined to quantify multi-dimensional priority characteristics (such as permission level and deadline), and a linear programming model is established based on resource pool constraints to achieve optimal initial resource allocation. The core technology lies in recursive causal analysis ensuring business process compliance, AHP empowerment enhancing decision interpretability, and constraint optimization improving resource utilization. Strong binding of permission tags and dependency paths prevents unauthorized operations; priority-driven allocation improves the resource satisfaction rate of high-value transactions; a dynamic normalization mechanism responds to permission changes in real time; and explicit weight matrices and causal graphs provide transparent evidence for auditing, ensuring strict compliance requirements and efficient resource allocation. S4: The correlation between the value and benefits of resource allocation in the transaction execution process of various government and enterprise project types is screened. A resource allocation benefit evaluation model for each government and enterprise project type is established, generating the optimal range of resource allocation-value benefit for each project type. The initial resource allocation for the transaction execution process of each government and enterprise project type is then corrected to obtain the optimal resource allocation scheme for each project type.

[0107] Step S4 specifically includes:

[0108] Obtain historical data on the execution progress and resource allocation of various government and enterprise projects.

[0109] Leveraging Bayesian networks, we can verify the posterior probability of the value and benefit impact of the transaction process request feature data of internal sub-nodes under the resource allocation of the transaction execution process for various types of government and enterprise projects, and determine the correlation matrix of resource allocation and value and benefit impact factors for the transaction execution process of various types of government and enterprise projects.

[0110] Using principal component analysis (PCA), the correlation matrix of resource allocation and value return influencing factors in the execution process of various government and enterprise projects is reduced in dimensionality to obtain the dimensionality-reduced correlation matrix of resource allocation and value return influencing factors in the execution process of various government and enterprise projects.

[0111] Based on random forest, a resource allocation benefit evaluation model for various types of government and enterprise projects is established, with the dimensionality reduction matrix of the correlation between resource allocation and value benefit influencing factors in the transaction execution process as the root node, the resource allocation and value benefit influencing factors in the transaction execution process as the branch nodes, and the optimal range of resource allocation and value benefit in the transaction execution process as the leaf nodes.

[0112] The initial resources of the transaction execution process of each type of government and enterprise project are used as the input of the resource allocation benefit evaluation model of each type of government and enterprise project. The constraint-sensitive split quality function is used as the split criterion to generate the optimal resource allocation scheme of each type of government and enterprise project as the output.

[0113] Specifically, the constraint-sensitive splitting quality function is:

[0114] ;

[0115] in, Let the terms be binary variables: the term for maximizing the purity of profit and the term for satisfying resource constraints. This represents the value corresponding to the transaction process request characteristic data of the k-th internal child node under the resource allocation of the transaction execution process for the i-th project type in historical government and enterprise projects. The average return on the execution process of various types of government and enterprise projects throughout history. A dimension-reduced matrix showing the correlation between resource allocation and value-benefit influencing factors in the execution process of various government and enterprise projects. This refers to the resource request vectors for the execution processes of various project types in government and enterprise sectors. The total amount of resources in the government and enterprise project pool for each project type. Assign constraint parameters to resources (assigning values ​​based on the priority of the transaction execution process). This represents the total number of internal child nodes.

[0116] When using it, please refer to the steps outlined above:

[0117] As a further step, this study utilizes Bayesian networks to uncover the causal relationship between historical resource allocation and value returns. Combined with PCA dimensionality reduction to extract core influencing factors, and employs random forest modeling to generate optimal resource-return intervals, the initial allocation scheme is finally revised based on constraint sensitivity criteria. The core technology lies in causal reasoning ensuring scientific decision-making, dimensionality reduction enhancing model generalization ability, and interval-based output increasing management flexibility. By eliminating subjective bias through causal relationships, the dimensionality-reduced model exhibits stronger noise resistance, the optimal interval provides flexible space for resource adjustments, and full-process automation ensures efficient resource allocation.

[0118] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for improving efficiency in digital management of government and enterprises, characterized in that, include: S1. Obtain multi-source heterogeneous data of various project types of government and enterprises, establish a business architecture knowledge graph of various project types of government and enterprises, perform semantic association according to the business architecture of various project types of government and enterprises, and generate a business architecture interaction data synchronization network of various project types of government and enterprises. S2. Based on the business architecture interaction data synchronization network of various government and enterprise project types, mark the execution process of transaction logs of various government and enterprise project types, establish a multi-level dynamic permission allocation model, and generate a dynamic permission network for business architecture interaction of various government and enterprise project types. S3. Based on the dynamic permission network of the business architecture interaction of various government and enterprise project types, verify the dependency relationship between each transaction execution process under the dynamic permission of the business architecture, establish a transaction execution process priority identification mechanism, assign priority to the transaction execution process, and determine the initial resource allocation of the transaction execution process of various government and enterprise project types. S4. Screen the correlation between the value and benefits of resource allocation in the execution process of various government and enterprise projects in history, establish a resource allocation benefit evaluation model for various government and enterprise projects, generate the optimal range of resource allocation-value benefit for the execution process of various government and enterprise projects, and revise the initial resource allocation for the execution process of various government and enterprise projects to obtain the optimal resource allocation scheme for each government and enterprise project type.

2. The efficiency improvement method for digital management of government and enterprises according to claim 1, characterized in that, Step S1 specifically includes: Based on multi-source heterogeneous data from various government and enterprise projects, the data is divided into structured, semi-structured, and unstructured data. A multi-head attention mechanism decoder is used to match the corresponding data types to generate a unified spatiotemporal reference data set for various government and enterprise project types. Based on a unified spatiotemporal reference dataset for various government and enterprise project types, the data is substituted into a BiLSTM-CRF joint relation extraction model. The BiLSTM is used to label the contextual semantic information of each structural data point within the unified spatiotemporal reference dataset for each government and enterprise project type at a given time, which is then input into the CRF. The conditional probabilities of entity / dependency labels corresponding to the contextual semantic information of each structural data point at a given time are calculated, thus determining the data entity-relationship probability distribution for each government and enterprise project type. The method is as follows: ； in, This refers to the conditional probabilities of entity / dependency labels given unified spatiotemporal benchmark data for various types of government and enterprise projects. A binary variable for entity / dependency label, This refers to the contextual semantic information of each structural data point in the unified spatiotemporal reference data for various government and enterprise project types at the t-th unit of time. This is a weighted matrix of entity / dependency labels. The entity / dependency label for the t-th unit of time is transferred to the t-th unit of time. Entity / dependency label offset per unit time, This refers to the sequence of all possible entity / dependency labels when providing unified spatiotemporal benchmark data for various government and enterprise project types.

3. The efficiency improvement method for digital management of government and enterprises according to claim 2, characterized in that, Step S1 also includes: Based on the data entity-relationship probability distribution of each type of government and enterprise project, and according to the business architecture level of each type of government and enterprise project, the data entity-relationship probability distribution of the corresponding business architecture level is substituted into the GNN graph neural network. The business architecture level is used as the global entity, the data entity of the corresponding type of government and enterprise project is used as the internal child node, and the data dependency relationship of the corresponding type of government and enterprise project is used as the edge weight to establish a data self-synchronization network of the business architecture of each type of government and enterprise project. Based on Jaccard-EditDistance hybrid similarity, this paper uses the Jaccard similarity algorithm to quantify the text character similarity of child nodes within the data self-synchronization network of business architectures for various government and enterprise project types. It then uses EditDistance to quantify the text character operation change similarity of these child nodes, determining the hybrid similarity between child nodes across different business architectures. This establishes an interactive data synchronization network for the business architectures of various government and enterprise project types, as follows: ； in, Let be the mixed similarity between the k-th child node and the (k+1)-th child node within the data self-synchronization network of the business architecture of the i-th project type in government and enterprise. These are the weighting coefficients. The text character similarity of the k-th child node within the data self-synchronization network of the business architecture for the i-th project type in government and enterprise. The similarity of text character operation changes within the k-th child node of the data self-synchronization network for the business architecture of the i-th project type in government and enterprise.

4. The efficiency improvement method for digital management of government and enterprises according to claim 3, characterized in that, Step S2 specifically includes: The execution process of the transaction logs of various government and enterprise project types is marked as the key-value attribute feature of the internal nodes in the business architecture interaction data synchronization network of various government and enterprise project types, thereby obtaining the sequence of sub-node transaction processes in the business architecture interaction data synchronization network of various government and enterprise project types. Based on the actual constraints of the business architecture of various government and enterprise projects, establish restrictions on transaction process state transitions. Based on the LSTM (Long Short-Term Memory) network, a multi-level dynamic permission allocation model is established. According to the transaction process state transition constraints, the interaction path constraints between the sub-nodes of the dynamic permission network for the interaction of business architectures of various government and enterprise project types are assigned. The transaction process sequence of the sub-nodes in the data synchronization network for the interaction of business architectures of various government and enterprise project types is used as input, and the data synchronization constraint network for the interaction of business architectures of various government and enterprise project types is used as output. Using NMTF nonnegative matrix trifactoring, we verify the strength of the dependency relationship between the transaction process pointing edges of internal child nodes in the network of business architecture interaction data synchronization constraints for various project types in government and enterprises. By utilizing the CVAE conditional variational autoencoder, the transaction process context information of internal child nodes in the business architecture interaction data synchronization constraint network for various government and enterprise project types is verified. The dependency strength of the transaction process pointing edges of internal child nodes is used as the latent vector. Using the reconstruction error and KL divergence, a permission allocation loss function is constructed to minimize the reconstruction error of the permission allocation loss function, the error between the actual permission allocation and the difference between the generated distribution and the prior distribution of KL divergence. Dynamic permissions are assigned to the transaction process sequences of child nodes in the business architecture interaction data synchronization network for various government and enterprise project types, resulting in a dynamic permission network for business architecture interaction for various government and enterprise project types.

5. The efficiency improvement method for digital management of government and enterprises according to claim 4, characterized in that, Step S3 specifically includes: Based on the dynamic permission network of business architecture interaction for various government and enterprise project types, obtain the transaction process and its dynamic permission tag in the business architecture of various government and enterprise project types. Based on recursive causal discovery, the causal relationship between the transaction process and dynamic permission tag of the internal child node in the dynamic permission network of the business architecture interaction of various project types of government and enterprise is calculated, the dependency path between the transaction processes of the internal child node is determined, and the dynamic permission dependency relationship network of the business architecture interaction of various project types of government and enterprise is obtained. Based on the dynamic permission dependency network of business architecture interaction for various government and enterprise project types, the transaction process request open metadata of internal sub-nodes under the dynamic permissions of business architecture interaction is obtained and normalized to obtain the transaction process request feature data of internal sub-nodes. Using the analytic hierarchy process (AHP), the strength of the dependency relationship between the transaction process pointing edges of the internal sub-nodes in the network is constrained according to the business architecture interaction data synchronization constraints of various government and enterprise project types, and the transaction process request feature data weights of the internal sub-nodes are assigned. The priority of the transaction process of internal sub-nodes in the network is determined by weighting and fusing the transaction process request feature data weights of internal sub-nodes with the transaction process request feature data of internal sub-nodes for business architecture interaction data synchronization constraints of various government and enterprise project types.

6. The efficiency improvement method for digital management of government and enterprises according to claim 5, characterized in that, Step S3 also includes: Based on the total amount of allocable resources per unit time for the business architecture of various government and enterprise project types, resource pool constraints are established for each government and enterprise project type. Based on the business architecture interaction data synchronization constraints of various government and enterprise project types, the transaction process priority vector of internal sub-nodes in the network is constructed. Normalize the transaction process priority vectors of internal child nodes; Determine the scope of resource requests for the transaction processes of internal sub-nodes in the network, constraining the business architecture interaction data synchronization constraints for various project types in government and enterprises. The objective function is to maximize the allocation of resource pools for each type of government and enterprise project to meet the business architecture interaction data synchronization constraints of each type of government and enterprise project. The resource pool constraints of each type of government and enterprise project are used as the total constraint, and the normalized value of the priority vector of the transaction process of the internal sub-nodes is used as the resource allocation decision variable to determine the initial resource allocation of the transaction execution process of each type of government and enterprise project.

7. The efficiency improvement method for digital management of government and enterprises according to claim 6, characterized in that, Step S4 specifically includes: Obtain historical data on the execution progress and resource allocation of various government and enterprise projects. Leveraging Bayesian networks, we can verify the posterior probability of the value and benefit impact of the transaction process request feature data of internal sub-nodes under the resource allocation of the transaction execution process for various types of government and enterprise projects, and determine the correlation matrix of resource allocation and value and benefit impact factors for the transaction execution process of various types of government and enterprise projects. Using principal component analysis (PCA), the correlation matrix of resource allocation and value return influencing factors in the execution process of various government and enterprise projects is reduced in dimensionality to obtain the dimensionality-reduced correlation matrix of resource allocation and value return influencing factors in the execution process of various government and enterprise projects. Based on random forest, a resource allocation benefit evaluation model for various types of government and enterprise projects is established, with the dimensionality reduction matrix of the correlation between resource allocation and value benefit influencing factors in the transaction execution process as the root node, the resource allocation and value benefit influencing factors in the transaction execution process as the branch nodes, and the optimal range of resource allocation and value benefit in the transaction execution process as the leaf nodes. The initial resources of the transaction execution process of each type of government and enterprise project are used as the input of the resource allocation benefit evaluation model of each type of government and enterprise project. The constraint-sensitive split quality function is used as the split criterion to generate the optimal resource allocation scheme of each type of government and enterprise project as the output. Specifically, the constraint-sensitive splitting quality function is: ； in, Let the terms be binary variables: the term for maximizing the purity of profit and the term for satisfying resource constraints. This represents the value corresponding to the transaction process request characteristic data of the k-th internal child node under the resource allocation of the transaction execution process for the i-th project type in historical government and enterprise projects. The average return on the execution process of various types of government and enterprise projects throughout history. A dimension-reduced matrix showing the correlation between resource allocation and value-benefit influencing factors in the execution process of various government and enterprise projects. This refers to the resource request vectors for the execution processes of various project types in government and enterprise sectors. The total amount of resources in the government and enterprise project pool for each project type. Resource allocation constraint parameters, This represents the total number of internal child nodes.

Images