A knowledge graph-based intelligent optimization system and method for enterprise management processes

Abstract

This application provides a knowledge graph-based intelligent optimization system and method for enterprise management processes. It maps the entity dependencies between various graph structure nodes in the enterprise management process to a knowledge graph, thereby generating a process knowledge graph for enterprise management. The system determines the temporal impact characteristics of content updates in the process knowledge graph through the dynamic decay factors of each management process subgraph. It calculates the data change amplitude and information entropy weight of each graph structure node, and then weights the data change amplitudes according to the information entropy weights to obtain the structural saliency of each management process subgraph. Based on the temporal impact characteristics and structural saliency, it determines the timeliness evaluation indicators for each graph structure node, and performs batch updates of the enterprise management process based on these timeliness evaluation indicators. This solution enables the integrated updating of data value and data timeliness in enterprise process management, thereby improving the execution efficiency of enterprise business processes.

Description

Technical Field

[0001] This application relates to the field of enterprise management technology, and more specifically, to an intelligent optimization system and method for enterprise management processes based on knowledge graphs. Background Technology

[0002] Enterprise management process optimization is developing rapidly, and enterprises are paying more and more attention to improving efficiency and competitiveness through process optimization. With the help of advanced information technology, enterprises can accurately analyze the bottleneck links in the process and realize intelligent process reengineering. Some small and medium-sized enterprises still face problems such as insufficient technology application capabilities and poor employee adaptability in the optimization process, but the optimization results can significantly improve the enterprise's operational efficiency and market response speed.

[0003] The use of static process templates in existing enterprise management processes has significant technical limitations, primarily in terms of dynamic adaptability. These templates typically operate based on pre-defined, fixed process logic, lacking the ability to perceive and respond to continuous changes in the business environment. When market strategies are adjusted, organizational structures are restructured, or business rules are updated, the system cannot autonomously identify these changes and make corresponding adjustments. This leads to a gradual disconnect between processes and actual business needs. The system's change mechanism often relies on manual intervention, requiring technical personnel to manually modify process definitions and redeploy, a process that is slow and prone to human error. Regarding process monitoring, traditional systems mainly focus on basic indicators of process completion, lacking the ability to dynamically evaluate process timeliness, compliance, and efficiency from multiple dimensions. Therefore, how to achieve the integrated updating of data value and data timeliness in enterprise process management to improve the execution efficiency of enterprise business processes has become a challenge for the industry. Summary of the Invention

[0004] This application provides a knowledge graph-based intelligent optimization system and method for enterprise management processes, which can integrate and update data value and data timeliness in enterprise process management, thereby improving the execution efficiency of enterprise business processes.

[0005] Firstly, this application provides a knowledge graph-based intelligent optimization method for enterprise management processes, including:

[0006] Generate a process knowledge graph for enterprise management based on structured and unstructured data modeling in multi-source heterogeneous enterprise systems;

[0007] The enterprise management process is decomposed into multiple graph structure nodes, and then the entity dependencies between the graph structure nodes are mapped into a knowledge graph, thereby generating a process knowledge graph for enterprise management.

[0008] Based on the business attributes in each graph structure node, a dynamic decay factor is set for each management process subgraph. Then, the temporal impact characteristics of content updates in the process knowledge graph are determined by the dynamic decay factor and the last update time of each management process subgraph in the process knowledge graph.

[0009] Calculate the data change magnitude and information entropy weight of each graph structure node, and then weight the data change magnitude and the maximum change value of data content in the enterprise management process according to the information entropy weight to obtain the structural saliency of each management process subgraph in the process knowledge graph.

[0010] Based on the time-series impact characteristics and the salience of each structure, the timeliness evaluation index of each graph structure node is determined, and the enterprise management process is updated in batches based on each timeliness evaluation index.

[0011] In some embodiments, decomposing enterprise management processes into multiple graph structure nodes specifically includes:

[0012] Extract multiple process elements from the enterprise management process from the enterprise management system documents, and then determine the business attribute tags of each process element;

[0013] Based on the various business attribute tags, each process element is structured into a directed graph, resulting in multiple graph structure nodes.

[0014] In some embodiments, mapping the entity dependencies between nodes of the graph structure to a knowledge graph, thereby generating a process knowledge graph for enterprise management, specifically includes:

[0015] Determine the entity dependencies between the nodes of each graph structure;

[0016] The enterprise management process knowledge graph is determined by all entity dependencies.

[0017] In some embodiments, determining the temporal impact characteristics of content updates in the process knowledge graph by using various dynamic decay factors and the last update time of each management process subgraph in the process knowledge graph specifically includes:

[0018] For each management process subgraph in the process knowledge graph, the update time difference of the management process subgraph is determined by the last update time of the management process subgraph and the current time.

[0019] Based on the update time difference and the dynamic decay factor of the management process subgraph, the sensitivity value of content update in the management process subgraph is determined, and then the sensitivity value of content update in each management process subgraph is obtained.

[0020] The temporal impact characteristics of content updates in the process knowledge graph are determined by all sensitivity values.

[0021] In some embodiments, calculating the data change magnitude and information entropy weight of each graph structure node specifically includes:

[0022] For each graph structure node, calculate the data change value of each numerical data in the graph structure node;

[0023] Determine the magnitude of data changes in the graph structure nodes by analyzing all the data change values.

[0024] The business rules based on graph structure nodes pre-set the information entropy weights of the graph structure nodes, thereby obtaining the data change magnitude and information entropy weights of each graph structure node.

[0025] In some embodiments, the structural saliency of each management process subgraph in the process knowledge graph is obtained by weighting the magnitude of each data change and the maximum change value of the data content in the enterprise management process according to the entropy weight of each information, specifically including:

[0026] Obtain the maximum change value of data content in the enterprise management process;

[0027] For each management process subgraph in the process knowledge graph, the significance value of the data content in the management process subgraph is determined based on the data change range of the corresponding graph structure node of the management process subgraph and the maximum change value.

[0028] The structural saliency of the management process subgraph is determined by the saliency value and the information entropy weight of the corresponding graph structure node of the management process subgraph, thereby obtaining the structural saliency of each management process subgraph in the process knowledge graph.

[0029] In some embodiments, determining the timeliness evaluation index for each graph structure node based on the time-series influence characteristics and the salience of each structure specifically includes:

[0030] For each graph structure node, industry adjustment coefficients for process nodes and data resources are set based on the industry characteristics of the graph structure node;

[0031] By using industry adjustment coefficients, the time-series impact characteristics and the structural salience of the management process subgraphs corresponding to the graph structure nodes are weighted according to time-sensitivity, and the time-sensitivity evaluation index of the graph structure nodes is obtained, thereby obtaining the time-sensitivity evaluation index of each graph structure node.

[0032] Secondly, this application provides a knowledge graph-based intelligent optimization system for enterprise management processes, including a batch update unit, wherein the batch update unit includes:

[0033] The graph generation module is used to generate a process knowledge graph for enterprise management based on structured and unstructured data modeling in a multi-source heterogeneous enterprise system.

[0034] The processing module is used to decompose the enterprise management process into multiple graph structure nodes, and then map the entity relationships between the graph structure nodes into a knowledge graph, thereby generating a process knowledge graph for enterprise management.

[0035] The processing module is also used to set the dynamic decay factor of each management process subgraph based on the business attributes in each graph structure node, and then determine the temporal impact characteristics of content updates in the process knowledge graph through each dynamic decay factor and the last update time of each management process subgraph in the process knowledge graph.

[0036] The processing module is also used to calculate the data change magnitude and information entropy weight of each graph structure node, and then to perform magnitude weighting on the data change magnitude and the maximum change value of the data content in the enterprise management process according to the information entropy weight, so as to obtain the structural saliency of each management process subgraph in the process knowledge graph.

[0037] The execution module is used to determine the timeliness evaluation index of each graph structure node based on the time sequence impact characteristics and the salience of each structure, and to perform batch updates of the enterprise management process based on each timeliness evaluation index.

[0038] Thirdly, this application provides a computer device, which includes a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-described intelligent optimization method for enterprise management processes based on knowledge graphs.

[0039] Fourthly, this application provides a computer-readable storage medium storing instructions or code that, when executed on a computer, cause the computer to implement the aforementioned knowledge graph-based intelligent optimization method for enterprise management processes.

[0040] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0041] This application provides a knowledge graph-based intelligent optimization system and method for enterprise management processes. It generates an enterprise management process knowledge graph based on structured and unstructured data modeling within a multi-source heterogeneous enterprise system. The enterprise management process is decomposed into multiple graph structure nodes, and the entity dependencies between these nodes are mapped to a knowledge graph, thus generating the enterprise management process knowledge graph. Dynamic decay factors are set for each management process subgraph based on the business attributes of each graph structure node. The temporal impact characteristics of content updates in the process knowledge graph are determined by using these dynamic decay factors and the last update time of each management process subgraph in the process knowledge graph. The data change amplitude and information entropy weight of each graph structure node are calculated. The amplitude of each data change and the maximum change value of the data content in the enterprise management process are then weighted according to the information entropy weight to obtain the structural saliency of each management process subgraph in the process knowledge graph. Based on the temporal impact characteristics and structural saliency, timeliness evaluation indicators for each graph structure node are determined, and the enterprise management process is updated in batches based on these timeliness evaluation indicators.

[0042] Therefore, this application determines the timeliness assessment indicators for each graph structure node based on the aforementioned time-series impact characteristics and the saliency of each structure. Based on these timeliness assessment indicators, batch updates of enterprise management processes are performed. First, determining the time-series impact characteristics yields the quantitative indicators of timeliness for each management process subgraph in the process knowledge graph, providing a scientific basis for enterprise process updates. By combining the time-series impact characteristics calculated with dynamic decay factors and the last update time, outdated process content can be accurately identified, thus enabling enterprises to distinguish between routine processes and time-sensitive processes. Processes with high timeliness requirements, such as financial approvals and market responses, can be prioritized for monitoring and rapid updates. When the time-series impact characteristics exceed a threshold, the system automatically triggers an update reminder or directly initiates the update process, ensuring that critical business processes always operate based on the latest policies and data, avoiding decision-making errors or execution deviations due to information lag. This dynamic timeliness management significantly improves the enterprise's ability to respond to market changes and regulatory requirements, thereby improving business process execution efficiency. Then, determining the structural saliency allows for the assessment of the importance of data changes in each subgraph of the process knowledge graph, achieving data-driven, precise updates. By calculating the significance of data changes based on both magnitude and value weight, the system can identify which data changes have the greatest impact on business. High-significance process sub-graphs often involve core business data or key decision-making nodes, and their updates will generate significant business value. The system intelligently allocates update resources based on significance levels: high-significance data changes are verified and updated immediately, medium-significance changes are included in periodic batch updates, and low-significance changes are simply recorded. This differentiated management ensures the timeliness of important data while avoiding unnecessary update overhead, thereby improving resource utilization for data updates and enhancing the accuracy of data in key business processes, significantly optimizing the quality of business process execution. In summary, based on the above solution, the integration of data value and data timeliness in enterprise process management can be achieved, thereby improving the execution efficiency of enterprise business processes. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is an exemplary flowchart of a knowledge graph-based intelligent optimization method for enterprise management processes, as shown in some embodiments of this application.

[0045] Figure 2This is a flowchart illustrating the process of determining structural saliency according to some embodiments of this application;

[0046] Figure 3 This is a schematic diagram of the structure of a batch update unit according to some embodiments of this application;

[0047] Figure 4 This is a schematic diagram of the structure of a computer device that implements a knowledge graph-based intelligent optimization method for enterprise management processes, according to some embodiments of this application. Detailed Implementation

[0048] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] refer to Figure 1 The figure is an exemplary flowchart of a knowledge graph-based intelligent optimization method for enterprise management processes, according to some embodiments of this application. This knowledge graph-based intelligent optimization method for enterprise management processes mainly includes the following steps:

[0050] In step 101, a process knowledge graph for enterprise management is generated based on structured and unstructured data modeling in the enterprise's multi-source heterogeneous system.

[0051] It should be noted that, in this application, a multi-source heterogeneous system refers to an information system that is deployed in a distributed manner within an enterprise, adopts different technical architectures, and has inconsistent data patterns; structured data is data with clearly defined formats and fields; unstructured data refers to data without a fixed format; and a process knowledge graph is a knowledge base that organizes and represents various entities, relationships, and rules in an enterprise's business processes in the form of a graph structure.

[0052] In practice, the process begins with extracting data from heterogeneous systems using a custom connector. Structured data is directly mapped to unified fields, while unstructured data is transformed into structured information using natural language processing techniques (e.g., entity recognition). Data cleaning includes deduplication, filling in missing values, and standardizing time formats. Finally, the structured and unstructured data are output as intermediate data with unified encoding and stored in a data warehouse. Next, for structured data, entity relationships are directly defined using foreign key constraints and business rules from a relational database. For unstructured text, pre-trained models are used to extract entities and relationships, such as identifying the "Party A-Party B-Contract Amount" triplet from contract text. This is combined with a domain dictionary to enhance accuracy, ultimately generating a preliminary graph framework containing nodes (entities) and edges (relationships). Finally, the merged entities and relationships are imported into a graph database. For example, nodes are set as "Department-Employee-Task," and edges are set as "Affiliation-Responsibility-Trigger." The graph pattern output from the graph database, conforming to the business scenario, serves as the process knowledge graph for enterprise management.

[0053] In step 102, the enterprise management process is decomposed into multiple graph structure nodes, and then the entity dependencies between the graph structure nodes are mapped into a knowledge graph, thereby generating a process knowledge graph for enterprise management.

[0054] In some embodiments, decomposing enterprise management processes into multiple graph structure nodes can be achieved using the following steps:

[0055] Extract multiple process elements from the enterprise management process from the enterprise management system documents, and then determine the business attribute tags of each process element;

[0056] Based on the various business attribute tags, each process element is structured into a directed graph, resulting in multiple graph structure nodes.

[0057] It should be noted that in this application, the graph structure node is the basic unit in the directed graph of enterprise management process; the process element is the basic unit reflecting the key entity actions in the business process; and the business attribute label is a classification identifier used to describe the characteristics of the process element.

[0058] In practice, the process begins by converting corporate management documents into processable text data using document parsing tools (e.g., PDF parsers). Irrelevant content is removed and the data is segmented by business modules. Natural language processing technology, combined with domain dictionaries and pre-trained models, is used to identify multiple process elements from the text, such as departmental approvals and budget applications. Structured information, such as "if {condition} then {action}", is extracted using rule templates. Attribute labels are then added to each process element based on the enterprise's business knowledge base. For example, "Approver" is categorized as "role-decision level" and "contract" is labeled as "form-legal document". Finally, the labeled process elements are transformed into graph structure nodes. Each graph structure node stores its attribute labels (e.g., type, constraints) and the original business description, resulting in multiple graph structure nodes.

[0059] In some embodiments, mapping the entity dependencies between nodes of a graph structure to a knowledge graph, thereby generating a process knowledge graph for enterprise management, can be achieved through the following steps:

[0060] Determine the entity dependencies between the nodes of each graph structure;

[0061] The enterprise management process knowledge graph is determined by all entity dependencies.

[0062] It should be noted that in this application, the process knowledge graph is a global graph model that integrates all nodes and their dependencies. This process knowledge graph can present the whole picture of the business process in a visual form; entity dependencies represent the business logic relationship between process elements.

[0063] In practical implementation, firstly, based on the constructed graph structure nodes, explicit and implicit dependencies between graph structure nodes are extracted by analyzing business rules and process documents. For explicit dependencies (such as approval chains explicitly stipulated in regulations), directed edges are generated by directly parsing sequential verbs in the text (such as "submitted to Y after review by X"). For implicit dependencies (such as data transfer in cross-departmental collaboration), potential relationships are mined by combining organizational structure and system logs (such as procurement applications needing to be synchronized to the financial system). The rationality of the dependencies is verified using a rule engine and graph algorithms (such as path analysis algorithms), ultimately forming a weighted graph structure. A set of directed relations, where weights reflect the strength of dependencies (e.g., mandatory processes and conditional branches), is used to define the set of explicit and implicit dependencies as entity dependencies, thus obtaining the entity dependencies between nodes in the graph structure. Then, a complete process knowledge graph is constructed using a graph database, and its correctness is verified through expert review or historical data to ensure the logical integrity of nodes and edges. For example, a reimbursement process must include the entire "application-approval-payment" chain, forming a structured knowledge network that supports process analysis, optimization, and intelligent applications. This structured knowledge network then serves as the process knowledge graph for enterprise management.

[0064] In step 103, dynamic decay factors are set for each management process subgraph based on the business attributes in each graph structure node. Then, the temporal impact characteristics of content updates in the process knowledge graph are determined by the dynamic decay factors and the last update time of each management process subgraph in the process knowledge graph.

[0065] In some embodiments, setting the dynamic decay factor of each management process subgraph based on the business attributes in each graph structure node can be achieved in the following way: Based on the existing node business attribute tags, the complete knowledge graph is divided into several management process subgraphs according to the business process domain. For each graph structure node, combined with business expert experience and historical process data, an initial dynamic decay factor is set for nodes of different attribute categories. Nodes with high timeliness are set with a higher dynamic decay factor, and basic nodes are set with a lower dynamic decay factor. By analyzing the correlation strength between the management process subgraphs corresponding to the graph structure nodes and the business process execution logs, the dynamic decay factor of the management process subgraph is dynamically adjusted. For example, the dynamic decay factor of the management process subgraph can be appropriately increased for process branches that have not been triggered for a long time, thereby obtaining the dynamic decay factor of each management process subgraph.

[0066] It should be noted that, in this application, the dynamic decay factor is a parameter used to quantify the degree of decay of the importance and correlation strength of each node in the process subgraph over time or business changes. This dynamic decay factor can reflect the timeliness characteristics of the node in the process. Business attribute classification refers to the classification of nodes in the process knowledge graph according to their business characteristics, such as: approval, execution and supervision. This business attribute classification can be used to distinguish business process nodes of different natures.

[0067] In some embodiments, determining the temporal impact characteristics of content updates in the process knowledge graph by using various dynamic decay factors and the last update time of each management process subgraph in the process knowledge graph can be achieved through the following steps:

[0068] For each management process subgraph in the process knowledge graph, the update time difference of the management process subgraph is determined by the last update time of the management process subgraph and the current time.

[0069] Based on the update time difference and the dynamic decay factor of the management process subgraph, the sensitivity value of content update in the management process subgraph is determined, and then the sensitivity value of content update in each management process subgraph is obtained.

[0070] The temporal impact characteristics of content updates in the process knowledge graph are determined by all sensitivity values.

[0071] It should be noted that, in this application, the temporal impact feature is an evaluation index characterizing the intensity of global update demand in the process knowledge graph; the last update time is the timestamp recording the most recent content change of each management process subgraph, and the last update time reflects the timeliness benchmark of the subgraph; the current time is the system time for performing the calculation, and this current time serves as a reference point for timeliness evaluation; the update time difference is the time interval that quantifies the obsolescence of the subgraph content; the dynamic decay factor is a parameter that determines the degree of influence of the time difference on the effectiveness of the content; and the sensitivity value is a quantitative index characterizing the urgency of content updates in the management process subgraph.

[0072] In practice, firstly, for each management process subgraph in the process knowledge graph, the difference between the last update time and the current time of the management process subgraph is taken as the update time difference of the management process subgraph; then, the product of the update time difference and the dynamic decay factor of the management process subgraph is taken as the sensitivity value of content update in the management process subgraph. The sensitivity value of content update in each management process subgraph can be obtained through the above method; finally, the sensitivity weight of each management process subgraph is obtained from the enterprise data governance specifications, and the weighted average of all sensitivity values ​​is calculated as the temporal influence feature of content update in the process knowledge graph.

[0073] In step 104, the data change magnitude and information entropy weight of each graph structure node are calculated. Then, based on the information entropy weight, the data change magnitude and the maximum change value of the data content in the enterprise management process are weighted by magnitude to obtain the structural saliency of each management process subgraph in the process knowledge graph.

[0074] In some embodiments, the calculation of the data variation magnitude and information entropy weight of each graph structure node can be achieved by the following steps:

[0075] For each graph structure node, calculate the data change value of each numerical data in the graph structure node;

[0076] Determine the magnitude of data changes in the graph structure nodes by analyzing all the data change values.

[0077] The business rules based on graph structure nodes pre-set the information entropy weights of the graph structure nodes, thereby obtaining the data change magnitude and information entropy weights of each graph structure node.

[0078] It should be noted that, in this application, the data change magnitude is an overall change intensity index that integrates the degree of change of all numerical attributes of a node; the information entropy weight is a value coefficient that reflects the criticality of the graph structure node data in the business process; numerical data refers to the business data in the node attributes that can be quantified and calculated; and the data change value is a quantitative index that reflects the degree of change of the node's numerical data within the current update cycle.

[0079] In practice, firstly, for each graph node, the difference between the current value and the last updated value of each numerical data in the graph node is calculated as the data change value of the corresponding numerical data, thus obtaining the data change value of each numerical data in the graph node; then, the calculation weights corresponding to each numerical data in the graph node are obtained from the enterprise data governance specifications, and a weighted average algorithm is used to calculate the weighted average of all data change values ​​as the data change amplitude of the graph node; finally, the information entropy weights of the graph node are obtained from the enterprise data governance specifications. Through the above methods, the data change amplitude and information entropy weight of each graph node can be obtained.

[0080] In some embodiments, the structural saliency of each management process subgraph in the process knowledge graph is obtained by weighting the magnitude of each data change and the maximum change value of the data content in the enterprise management process according to the entropy weight of each information, and then referring to... Figure 2 The diagram is a flowchart illustrating the determination of structural saliency in some embodiments of this application. In this embodiment, the determination of structural saliency can be achieved using the following steps:

[0081] In step 1041, the maximum change value of the data content in the enterprise management process is obtained;

[0082] In step 1042, for each management process subgraph in the process knowledge graph, the significance value of the data content in the management process subgraph is determined based on the data change range of the corresponding graph structure node of the management process subgraph and the maximum change value.

[0083] In step 1043, the structural salience of the management process subgraph is determined by the salience value and the information entropy weight of the graph structure node corresponding to the management process subgraph, thereby obtaining the structural salience of each management process subgraph in the process knowledge graph.

[0084] It should be noted that, in this application, structural saliency is a quantitative value that characterizes the actual impact of the management process subgraph on enterprise operations; maximum variation value is a benchmark reference value for measuring the degree of data change; and significance value is an indicator that reflects the degree of significance of data changes within the management process subgraph.

[0085] In practice, firstly, the maximum value among all data variation ranges is taken as the maximum variation value of the data content in the enterprise management process; then, for each management process subgraph in the process knowledge graph, the ratio of the data variation range of the corresponding graph structure node of the management process subgraph to the maximum variation value is taken as the saliency value of the data content in the management process subgraph; finally, the product of the saliency value and the information entropy weight of the corresponding graph structure node of the management process subgraph is taken as the structural saliency of the management process subgraph. The structural saliency of each management process subgraph in the process knowledge graph can be obtained through the above method.

[0086] In step 105, the timeliness evaluation index of each graph structure node is determined based on the time sequence impact characteristics and the salience of each structure, and the enterprise management process is updated in batches based on each timeliness evaluation index.

[0087] In some embodiments, determining the timeliness evaluation index of each graph structure node based on the time-series influence characteristics and the salience of each structure can be achieved by the following steps:

[0088] For each graph structure node, industry adjustment coefficients for process nodes and data resources are set based on the industry characteristics of the graph structure node;

[0089] By using industry adjustment coefficients, the time-series impact characteristics and the structural salience of the management process subgraphs corresponding to the graph structure nodes are weighted according to time-sensitivity, and the time-sensitivity evaluation index of the graph structure nodes is obtained, thereby obtaining the time-sensitivity evaluation index of each graph structure node.

[0090] It should be noted that in this application, the timeliness assessment index is a quantitative measure of the update and maintenance priority of nodes within an industry context. Specifically, for each graph structure node, firstly, the industry characteristics of the graph structure node are obtained. The adjustment weights of process nodes and data resources within these industry characteristics are obtained from the enterprise data governance specifications as industry adjustment coefficients. These industry adjustment coefficients reflect the degree of influence of industry differences on timeliness assessment. Then, the industry adjustment coefficients of process nodes are used as the calculation weights for time-series impact characteristics, and the industry adjustment coefficients of data resources are used as the calculation weights for structural saliency. The weighted sum of time-series impact characteristics and structural saliency is then calculated as the timeliness assessment index for the graph structure node. The timeliness assessment index for each graph structure node can be obtained through the above method.

[0091] In some embodiments, batch updates of enterprise management processes based on various timeliness assessment indicators can be implemented in the following way: Establish update batch division rules, dividing nodes into four batches according to timeliness assessment indicators: immediate update, same-day update, periodic update, and passive update. Design differentiated update strategies for different batches. Immediate update batches trigger real-time alarms and automatically lock related business systems, and are handled by designated personnel with priority. Same-day update batches are included in the daily maintenance work order system and processed centrally during off-peak business periods. Periodic update batches are executed in conjunction with the enterprise's monthly / quarterly maintenance plan. Passive update batches are only triggered when actual business changes occur. In the specific implementation process, update logs are recorded through a version control system, and a dual verification mechanism is set to ensure the accuracy of updates for high-level nodes. After each batch update, the timeliness assessment indicators of relevant nodes are automatically recalculated, forming a closed-loop optimization system. This closed-loop optimization system is driven by a workflow engine and deeply integrated with the enterprise process management system to achieve full-link automated management from assessment to update.

[0092] Furthermore, in another aspect of this application, in some embodiments, this application provides a knowledge graph-based intelligent optimization system for enterprise management processes. This knowledge graph-based intelligent optimization system for enterprise management processes includes a batch update unit, referencing... Figure 3 The figure is a schematic diagram of the structure of a batch update unit according to some embodiments of this application. The batch update unit includes: a map generation module 201, a processing module 202, and an execution module 203, which are described below:

[0093] Knowledge graph generation module 201 in this application is mainly used to generate a process knowledge graph for enterprise management based on structured and unstructured data modeling in a multi-source heterogeneous system of an enterprise.

[0094] Processing module 202, in this application, is used to decompose the enterprise management process into multiple graph structure nodes, and then map the entity dependency relationship between each graph structure node into a knowledge graph, thereby generating a process knowledge graph of enterprise management.

[0095] It should be noted that the processing module 202 is also used to set the dynamic decay factor of each management process subgraph based on the business attributes in each graph structure node, and then determine the temporal impact characteristics of content updates in the process knowledge graph through each dynamic decay factor and the last update time of each management process subgraph in the process knowledge graph.

[0096] In addition, the processing module 202 is also used to calculate the data change magnitude and information entropy weight of each graph structure node, and then to perform magnitude weighting on each data change magnitude and the maximum change value of data content in the enterprise management process according to each information entropy weight, so as to obtain the structural saliency of each management process subgraph in the process knowledge graph.

[0097] The execution module 203 in this application is mainly used to determine the timeliness evaluation index of each graph structure node based on the time sequence influence characteristics and the salience of each structure, and to perform batch updates of the enterprise management process based on each timeliness evaluation index.

[0098] The foregoing has detailed examples of the knowledge graph-based intelligent optimization system and method for enterprise management processes provided in this application. It is understood that the corresponding apparatus, in order to achieve the above functions, includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0099] In some embodiments, this application also provides a computer device, the computer device including a memory and a processor, the memory for storing a computer program, and the processor for calling and running the computer program from the memory, so that the computer device executes the above-described intelligent optimization method for enterprise management processes based on knowledge graphs.

[0100] In some embodiments, reference Figure 4The dashed lines in the figure indicate that the unit or module is optional. This figure is a structural schematic diagram of a computer device implementing a knowledge graph-based intelligent optimization method for enterprise management processes according to an embodiment of this application. The knowledge graph-based intelligent optimization method for enterprise management processes described in the above embodiments can... Figure 4 The computer device shown is used to implement this, and the computer device includes at least one processor 301, a memory 302 and at least one communication unit 305. The computer device may be a terminal device, a server or a chip.

[0101] Processor 301 can be a general-purpose processor or a special-purpose processor. For example, processor 301 can be a central processing unit (CPU), which can be used to control computer devices, execute software programs, and process data from software programs. The computer device may also include a communication unit 305 for inputting (receiving) and outputting (transmitting) signals.

[0102] For example, the computer device may be a chip, and the communication unit 305 may be the input and / or output circuit of the chip, or the communication unit 305 may be the communication interface of the chip, which may be a component of a terminal device, network device or other device.

[0103] For example, the computer device may be a terminal device or a server, and the communication unit 305 may be a transceiver of the terminal device or the server, or the communication unit 305 may be a transceiver circuit of the terminal device or the server.

[0104] The computer device may include one or more memories 302 storing a program 304. The program 304 can be executed by a processor 301 to generate instructions 303, causing the processor 301 to execute the method described in the above method embodiments according to the instructions 303. Optionally, the memory 302 may also store data (such as a target audit model). Optionally, the processor 301 may also read data stored in the memory 302, which may be stored at the same storage address as the program 304, or it may be stored at a different storage address than the program 304.

[0105] The processor 301 and memory 302 can be configured separately or integrated together, for example, integrated on the system on chip (SOC) of the terminal device.

[0106] It should be understood that each step of the above method embodiment can be completed by hardware logic circuits or software instructions in the processor 301. The processor 301 can be a CPU, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, such as discrete gate, transistor logic devices, or discrete hardware components.

[0107] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] For example, in some embodiments, this application also provides a computer-readable storage medium storing instructions or code that, when executed on a computer, cause the computer to implement the above-described intelligent optimization method for enterprise management processes based on knowledge graphs.

[0109] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0110] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A knowledge graph-based intelligent optimization method for enterprise management processes, characterized in that, Includes the following steps: Generate a process knowledge graph for enterprise management based on structured and unstructured data modeling in multi-source heterogeneous enterprise systems; The enterprise management process is decomposed into multiple graph structure nodes, and then the entity dependencies between the graph structure nodes are mapped into a knowledge graph, thereby generating a process knowledge graph for enterprise management. Based on the business attributes in each graph structure node, a dynamic decay factor is set for each management process subgraph. Then, the temporal impact characteristics of content updates in the process knowledge graph are determined by the dynamic decay factor and the last update time of each management process subgraph in the process knowledge graph. Calculate the data change magnitude and information entropy weight of each graph structure node, and then weight the data change magnitude and the maximum change value of data content in the enterprise management process according to the information entropy weight to obtain the structural saliency of each management process subgraph in the process knowledge graph. Based on the time-series impact characteristics and the salience of each structure, the timeliness evaluation index of each graph structure node is determined, and the enterprise management process is updated in batches based on each timeliness evaluation index. Specifically, determining the temporal impact characteristics of content updates in the process knowledge graph by using various dynamic decay factors and the last update time of each management process subgraph in the process knowledge graph includes: For each management process subgraph in the process knowledge graph, the update time difference of the management process subgraph is determined by the last update time of the management process subgraph and the current time. Based on the update time difference and the dynamic decay factor of the management process subgraph, the sensitivity value of content update in the management process subgraph is determined, and then the sensitivity value of content update in each management process subgraph is obtained. Determine the temporal impact characteristics of content updates in the process knowledge graph by identifying all sensitive values; Specifically, the structural saliency of each management process subgraph in the process knowledge graph is obtained by weighting the magnitude of changes in each data point and the maximum change in data content within the enterprise management process according to the entropy weights of each information. This includes: Obtain the maximum change value of data content in the enterprise management process; For each management process subgraph in the process knowledge graph, the significance value of the data content in the management process subgraph is determined based on the data change range of the corresponding graph structure node of the management process subgraph and the maximum change value. The structural saliency of the management process subgraph is determined by the saliency value and the information entropy weight of the corresponding graph structure node of the management process subgraph, thereby obtaining the structural saliency of each management process subgraph in the process knowledge graph.

2. The method as described in claim 1, characterized in that, Decomposing enterprise management processes into multiple graph structure nodes specifically includes: Extract multiple process elements from the enterprise management process from the enterprise management system documents, and then determine the business attribute tags of each process element; Based on the various business attribute tags, each process element is structured into a directed graph, resulting in multiple graph structure nodes.

3. The method as described in claim 1, characterized in that, Mapping the entity dependencies between nodes in the graph structure to a knowledge graph, and then generating a process knowledge graph for enterprise management, specifically includes: Determine the entity dependencies between the nodes of each graph structure; The enterprise management process knowledge graph is determined by all entity dependencies.

4. The method as described in claim 1, characterized in that, The calculation of the data variation magnitude and information entropy weight of each graph structure node specifically includes: For each graph structure node, calculate the data change value of each numerical data in the graph structure node; Determine the magnitude of data changes in the graph structure nodes by analyzing all the data change values. The business rules based on graph structure nodes pre-set the information entropy weights of the graph structure nodes, thereby obtaining the data change magnitude and information entropy weights of each graph structure node.

5. The method as described in claim 1, characterized in that, Based on the aforementioned temporal impact characteristics and the salience of each structure, the timeliness evaluation indicators for each graph structure node are specifically determined as follows: For each graph structure node, industry adjustment coefficients for process nodes and data resources are set based on the industry characteristics of the graph structure node; By using industry adjustment coefficients, the time-series impact characteristics and the structural salience of the management process subgraphs corresponding to the graph structure nodes are weighted according to time-sensitivity, and the time-sensitivity evaluation index of the graph structure nodes is obtained, thereby obtaining the time-sensitivity evaluation index of each graph structure node.

6. A knowledge graph-based intelligent optimization system for enterprise management processes, comprising a batch update unit, wherein the system uses the method described in any one of claims 1 to 5 to perform knowledge graph-based intelligent optimization of enterprise management processes, and the knowledge graph-based intelligent optimization system for enterprise management processes comprises a batch update unit, characterized in that... The batch update unit includes: The graph generation module is used to generate a process knowledge graph for enterprise management based on structured and unstructured data modeling in a multi-source heterogeneous enterprise system. The processing module is used to decompose the enterprise management process into multiple graph structure nodes, and then map the entity relationships between the graph structure nodes into a knowledge graph, thereby generating a process knowledge graph for enterprise management. The processing module is also used to set the dynamic decay factor of each management process subgraph based on the business attributes in each graph structure node, and then determine the temporal impact characteristics of content updates in the process knowledge graph through each dynamic decay factor and the last update time of each management process subgraph in the process knowledge graph. The processing module is also used to calculate the data change magnitude and information entropy weight of each graph structure node, and then to perform magnitude weighting on the data change magnitude and the maximum change value of the data content in the enterprise management process according to the information entropy weight, so as to obtain the structural saliency of each management process subgraph in the process knowledge graph. The execution module is used to determine the timeliness evaluation index of each graph structure node based on the time sequence impact characteristics and the salience of each structure, and to perform batch updates of the enterprise management process based on each timeliness evaluation index.

7. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to call and run the computer programs from the memory, so that the computer device executes the knowledge graph-based intelligent optimization method for enterprise management processes as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions or code that, when executed on a computer, cause the computer to implement the knowledge graph-based intelligent optimization method for enterprise management processes as described in any one of claims 1 to 5.

Images