Enterprise digital management method and system based on data mining

Abstract

The invention discloses an enterprise digital management method based on data mining. The method comprises the following steps: acquiring information flow data and resource allocation data; analyzing according to the information flow data and the resource allocation data to obtain an initial network diagram; performing clustering analysis according to the initial network diagram, and performing path optimization to obtain a dynamic network diagram; performing feature extraction on the dynamic network diagram to obtain prediction analysis data; according to the dynamic network diagram and the prediction analysis data, network nodes are identified, and key nodes and high-risk nodes are obtained; in combination with the key nodes and the high-risk nodes, identifying a key propagation path to obtain the key propagation path; and based on the key nodes, the high-risk nodes and the key propagation paths, predicting key positions where resource bottlenecks and conflicts occur, and generating resource optimization suggestions. According to the method, the dynamic change of the implicit dependency relationship of the enterprise can be identified, high-risk nodes and key propagation paths are positioned, and resource configuration is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise digital management, and particularly to an enterprise digital management method and system based on data mining. Background Art

[0002] At present, in the process of enterprise digital management, traditional management methods rely on static organizational charts, process management systems (BPM), or rule-based resource allocation schemes to describe the collaboration patterns between departments. However, these methods cannot accurately reflect the dynamic changes in the internal resource flow of enterprises and are also difficult to adapt to the rapid adjustment of the enterprise environment.

[0003] In an existing technology, methods such as graph theory analysis or social network analysis (SNA) are used to model the internal dependencies of enterprises. For example, a network diagram is constructed to describe the resource allocation pattern between departments, and centrality indicators are used to analyze key nodes. However, the existing technology is mainly based on a static network model and is difficult to capture the dynamic changes in the internal dependencies of enterprises, resulting in low real-time performance and adaptability of the analysis results.

[0004] The existing technology cannot accurately identify and analyze the dynamic changes in the internal implicit dependencies of enterprises, resulting in inaccurate positioning of high-risk nodes and key propagation paths, which affects the resource optimization of enterprises. Summary of the Invention

[0005] The present invention provides an enterprise digital management method and system based on data mining. The present invention can identify the dynamic changes in the implicit dependencies of enterprises, locate high-risk nodes and key propagation paths, and optimize resource allocation.

[0006] In a first aspect, to solve the above technical problems, the present invention provides an enterprise digital management method based on data mining, including:

[0007] Obtaining the information flow data and resource allocation data between departments;

[0008] Performing dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of the dependencies between departments;

[0009] Based on the initial network diagram, performing clustering analysis on the connection strength and path length, dynamically reconstructing the network, and fusing community mining and path optimization to identify key nodes to obtain a dynamic network diagram of the information flow between departments;

[0010] Performing non-linear feature extraction of the interaction effect on the dynamic network diagram to obtain predictive analysis data for judging the node influence;

[0011] Based on the dynamic network diagram and the predictive analysis data, identify the nodes in the network that affect resource allocation and information transmission to obtain key nodes and high-risk nodes;

[0012] Based on the key nodes and the high-risk nodes, identify the key propagation paths in the network to obtain the key propagation paths that affect resource allocation and information transmission;

[0013] Based on the key nodes, the high-risk nodes and the key propagation paths, predict the key locations where resource bottlenecks and resource conflicts occur, and generate resource optimization suggestions.

[0014] Preferably, based on the information flow data and the resource allocation data, perform dependency relationship analysis to obtain an initial network diagram of inter-departmental dependencies, including:

[0015] Calculate the support degree and confidence degree between departments according to the information flow data and the resource allocation data;

[0016] When the support degree and the confidence degree are respectively greater than the preset support degree threshold and the preset confidence degree threshold, based on graph theory methods, establish a network structure diagram of inter-departmental dependencies to obtain an initial network diagram of inter-departmental dependencies.

[0017] Preferably, based on the initial network diagram, perform clustering analysis on the connection strength and path length, and dynamically reconstruct the network and fuse community mining and path optimization to identify key nodes to obtain a dynamic network diagram of inter-departmental information flow, including:

[0018] Extract the connection strength data and path length data from the initial network diagram;

[0019] Perform resource allocation pattern analysis according to the connection strength data and the path length data to obtain the main transmission paths of inter-departmental dependencies;

[0020] Perform dependency trend analysis on the main transmission paths to obtain the inter-departmental dependency trend;

[0021] When the dependency trend is greater than the preset trend range, update the network structure of the initial network diagram based on graph theory methods to obtain an updated network diagram of inter-departmental dependencies;

[0022] Identify the key nodes for resource allocation according to the updated network diagram to obtain the priority nodes that require priority resource allocation;

[0023] Perform path weight analysis on the priority nodes to obtain a weight factor reflecting the importance of inter-departmental information transmission;

[0024] Perform an operation to update the dependencies between departments according to the weight factors, and obtain a dynamic network diagram of information flow between departments.

[0025] Preferably, perform non-linear feature extraction on the dynamic network diagram to obtain predictive analysis data for judging node influence, including:

[0026] Extract the node connection relationships, weight data, and time series data in the dynamic network diagram;

[0027] Calculate the influence index of each node according to the node connection relationship and the weight data;

[0028] When the influence index is greater than the preset influence threshold, determine the node corresponding to the influence index as the main node with an amplification effect;

[0029] When the influence index is less than the preset influence threshold, determine the node corresponding to the influence index as the secondary node with a weakening effect;

[0030] Based on the time series data, perform an evolutionary trend analysis on the main nodes and secondary nodes to obtain predictive analysis data on the evolutionary trend of node influence.

[0031] Preferably, according to the dynamic network diagram and the predictive analysis data, identify the nodes in the network that affect resource allocation and information transmission, and obtain key nodes and high-risk nodes, including:

[0032] Extract the node connection data, shortest path data, network topology change trend, clustering coefficient, and community membership data in the dynamic network diagram;

[0033] Based on the predictive analysis data, combined with the node connection data, the shortest path data, and the network topology change trend, perform an operation to correct the dynamic weight of the centrality index to obtain the centrality index for evaluating the importance of nodes in information dissemination;

[0034] Based on the predictive analysis data, combined with the clustering coefficient and the community membership data, perform a correction of the dynamic weight of the vulnerability index to obtain the vulnerability index for evaluating the dependence degree of nodes in information dissemination;

[0035] According to the centrality index, construct the feature vectors of each node to obtain the centrality feature vector;

[0036] Use the K-means clustering algorithm to perform importance ranking and screening on the centrality feature vector to obtain preliminary key nodes with high importance, medium importance, and low importance;

[0037] Rank the importance of the preliminary key nodes to obtain key nodes with high centrality;

[0038] Analyze the connection characteristics and load capacity indicators of each node according to the vulnerability indicators to obtain a vulnerability evaluation value;

[0039] When the vulnerability evaluation value is less than the preset vulnerability threshold, determine the node corresponding to the vulnerability evaluation value as a potential high-risk node;

[0040] Use the support vector machine algorithm to predict node failures for the potential high-risk nodes to obtain node failure probabilities;

[0041] Extract the set of nodes whose node failure probability is greater than the preset probability threshold, and determine all nodes in the node set as high-risk nodes.

[0042] Preferably, the identifying the critical propagation paths in the network according to the key nodes and the high-risk nodes to obtain the critical propagation paths affecting resource allocation and information transmission includes:

[0043] Construct a risk network graph using graph theory methods according to the key nodes and the high-risk nodes;

[0044] Calculate the path parameters of the risk network graph to obtain the propagation data and stability data of the nodes on the path;

[0045] Evaluate the node importance and path stability according to the propagation data and stability data to obtain a path evaluation result;

[0046] Group the path evaluation results according to the path propagation priority, resource influence, and decision chain weight to obtain the critical propagation paths affecting resource allocation and information transmission.

[0047] Preferably, the predicting the critical positions where resource bottlenecks and resource conflicts occur based on the key nodes, the high-risk nodes, and the critical propagation paths, and generating resource optimization suggestions includes:

[0048] Analyze the resource allocation status according to the key nodes, the high-risk nodes, and the critical propagation paths to obtain the resource usage of the key nodes, the resource pressure of the high-risk nodes, and the traffic load of the critical propagation paths;

[0049] Predict the critical positions where resource bottlenecks and resource conflicts occur according to the resource usage, the resource pressure, and the traffic load, in combination with the preset historical resource scheduling data;

[0050] Adjust the resource allocation strategy for the critical positions to generate resource optimization suggestions.

[0051] In a second aspect, the present invention provides an enterprise digital management system based on data mining, including:

[0052] A data acquisition module for acquiring information flow data and resource allocation data between departments;

[0053] A dependency analysis module for performing dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of the inter-departmental dependency relationship;

[0054] A path optimization module for performing clustering analysis on connection strength and path length based on the initial network diagram, dynamically reconstructing the network, and fusing community mining and path optimization to identify key nodes, thereby obtaining a dynamic network diagram of information flow between departments;

[0055] A prediction analysis module for extracting non-linear features of interaction effects from the dynamic network diagram to obtain prediction analysis data for judging node influence;

[0056] A node identification module for identifying nodes in the network that affect resource allocation and information transmission based on the dynamic network diagram and the prediction analysis data to obtain key nodes and high-risk nodes;

[0057] A critical path module for identifying critical propagation paths in the network based on the key nodes and the high-risk nodes to obtain critical propagation paths that affect resource allocation and information transmission;

[0058] A resource optimization module for predicting critical positions where resource bottlenecks and resource conflicts occur based on the key nodes, the high-risk nodes, and the critical propagation paths, and generating resource optimization suggestions.

[0059] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the above-mentioned enterprise digital management method based on data mining is implemented.

[0060] In a fourth aspect, the present invention further provides a computer-readable storage medium. The computer-readable storage medium includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the above-mentioned enterprise digital management method based on data mining.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] (1)The present invention provides a method for enterprise digital management based on data mining. By obtaining the information flow data and resource allocation data among enterprise departments, an initial network diagram is constructed, and combined with time series analysis and path optimization algorithms, the enterprise dependency network is dynamically adjusted. The prior art mainly relies on static analysis methods, which are difficult to accurately reflect the dynamic evolution of internal dependencies within an enterprise, resulting in lagging resource allocation decisions. The present invention can update the network structure in real time and identify the changing trends of internal information flows and resource flows within the enterprise.

[0063] (2)The present invention adopts a non - linear regression analysis method, combined with centrality indicators and vulnerability indicators, to identify high - risk nodes and key propagation paths. The prior art is based on linear analysis and is difficult to capture the non - linear changes in the influence of certain nodes under different business scenarios. The present invention calculates the node influence index, combines the support vector machine algorithm to predict the node failure probability, and accurately identifies the nodes that have a greater impact on enterprise stability. At the same time, based on the path analysis algorithm, key propagation paths are discovered, revealing the core links within the enterprise that lead to business interruptions or resource conflicts.

[0064] (3)The present invention is based on path analysis and machine learning algorithms to predict resource bottlenecks and resource conflicts and generate optimization suggestions. Compared with the prior art methods that rely on fixed rules for resource management, the present invention dynamically calculates the resource usage, traffic load, and critical path dependency, and combines optimization algorithms to adjust the resource allocation strategy, making resource scheduling more flexible and efficient. At the same time, the optimized dynamic network diagram can be integrated with the enterprise resource management system to achieve real - time monitoring of dependencies between departments and the status of resource flows, improve the enterprise's response ability in complex business environments, and optimize the resource allocation plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is a schematic flow chart of the method for enterprise digital management based on data mining provided by the first embodiment of the present invention;

[0066] Figure 2 is a schematic structural diagram of the enterprise digital management system based on data mining provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0068] Refer to Figure 1, the first embodiment of the present invention provides an enterprise digital management method based on data mining, including the following steps:

[0069] S11, obtain information flow data and resource allocation data between departments;

[0070] S12, perform dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of the dependency relationship between departments;

[0071] S13, based on the initial network diagram, perform clustering analysis on connection strength and path length, and dynamically reconstruct the network and fuse community mining and path optimization to identify key nodes, obtaining a dynamic network diagram of information flow between departments;

[0072] S14, extract non-linear features of interaction effects from the dynamic network diagram to obtain predictive analysis data for judging node influence;

[0073] S15, based on the dynamic network diagram and the predictive analysis data, identify nodes in the network that affect resource allocation and information transmission, obtaining key nodes and high-risk nodes;

[0074] S16, based on the key nodes and the high-risk nodes, identify key propagation paths in the network, obtaining key propagation paths that affect resource allocation and information transmission;

[0075] S17, based on the key nodes, the high-risk nodes, and the key propagation paths, predict the key positions where resource bottlenecks and resource conflicts occur, and generate resource optimization suggestions.

[0076] In step S11, it is necessary to obtain information flow data and resource allocation data between departments, including:

[0077] In a specific embodiment, the information flow data mainly comes from the enterprise's internal office system, communication system, business management system, and data interaction platform. For example, the document management system (DMS), enterprise email system (Email), and instant messaging tools (such as enterprise WeChat, DingTalk) in the office system can provide data such as file sharing, email exchanges, and communication records. The data of the communication system includes telephone conference records, video conference logs, enterprise internal forums, and work logs, etc., which are used to reflect the communication frequency and information exchange content between different departments. In addition, the enterprise's business management systems (such as ERP, CRM, SCM, etc.) can record cross-departmental business collaboration relationships, such as order flow, customer demand feedback, production scheduling, etc. The data interaction platform (such as the enterprise data sharing platform and API logs) can provide the data call frequency and data sharing situation between different systems, further revealing the information dependency relationship between departments.

[0078] In a specific embodiment, the resource allocation data mainly involves aspects such as human resources within the enterprise, capital flow, equipment usage, and production resource allocation. Among them, human resource data can be obtained through the enterprise's Human Resource Management System (HRM), including the personnel distribution in each department, job transfers, and the man-hour allocation for cross-departmental collaboration. The Financial Management System (FMS) provides information on capital flow, such as budget allocation, project costs, and cross-departmental fund transactions. The Asset Management System (AMS) is used to record the usage of enterprise equipment, office resources, and production tools, and track the sharing and allocation of equipment among different departments. The Production and Supply Chain Management System (SCM) reflects the inventory turnover, production material scheduling, and logistics distribution, ensuring that the resource allocation data comprehensively covers all aspects of enterprise operations.

[0079] Specifically, to ensure the integrity and timeliness of the data, the present invention adopts various data collection methods, including system log analysis, database query, API data call, and manual input review. System log analysis mainly targets the data of office systems and communication systems, automatically extracting email logs, file sharing records, and meeting communication data. For example, by parsing the email server logs, the cross-departmental email exchanges can be obtained, and the content relevance can be analyzed by combining the email subject and keywords. The database query method is applicable to business management systems, and the internal resource flow data of the enterprise, such as the approval process of purchase orders and fund payment records, can be directly extracted from databases such as ERP and CRM through SQL queries. The API data call method is applicable to the real-time data collection of different business systems. For example, the interfaces of the ERP and SCM systems are called to obtain inventory change information and material allocation situations. In addition, for some unstructured data, such as the resource coordination meeting records between departments, the data integrity can be ensured through manual review and supplementation.

[0080] It should be noted that since the collected data comes from diverse sources and has different formats, the present invention uses methods of data cleaning, normalization, and format conversion for preprocessing. Data cleaning mainly removes duplicate, invalid, and abnormal data. For example, auto-reply emails are excluded to avoid interfering with the information flow analysis. Data normalization ensures that data from different sources have a unified measurement standard. For example, the man-hour data recorded in different systems is converted into standard hour units for cross-departmental comparison. Format conversion is used to convert unstructured data (such as meeting records, text emails) into structured data tables for storage. For example, through natural language processing (NLP) technology, the keywords of the email content are extracted and associated with the corresponding business processes.

[0081] Taking a manufacturing enterprise as an example, the dependency relationships between its R & D department and production department are mainly reflected in the interaction of technical documents, the sharing of experimental equipment, and the arrangement of testing personnel. The present invention can extract the design documents uploaded by the R & D department through the DMS system, analyze their transfer paths among different departments to identify the dissemination patterns of technical knowledge. At the same time, the usage logs of experimental equipment are obtained through the Asset Management System (AMS) to track the sharing of equipment between the R & D and production departments, so as to analyze the cross-departmental flow of resources. The Human Resource Management System (HRM) provides the work scheduling data of testing personnel and analyzes the man-hour allocation of testing personnel in different projects to judge the dependency relationships of cross-departmental human resources. In the data preprocessing stage, all data are normalized according to a unified time dimension for subsequent time series analysis and trend prediction.

[0082] In step S12, it is necessary to perform dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of inter-departmental dependencies, including:

[0083] Calculate the support degree and confidence degree between departments according to the information flow data and the resource allocation data;

[0084] When the support degree and the confidence degree are respectively greater than the preset support degree threshold and the preset confidence degree threshold, based on the graph theory method, establish a network structure diagram of inter-departmental dependencies to obtain an initial network diagram of inter-departmental dependencies.

[0085] First of all, in order to ensure the accuracy of the dependency relationship analysis, the present invention performs data fusion on the information flow data and the resource allocation data. The main purpose of data fusion is to integrate multiple data sources, make up for the deficiencies of a single data source, and improve the integrity and consistency of data. Data fusion adopts methods such as time alignment, entity matching, and similarity analysis. For example, for the data of the office system, the business management system, and the human resource system, first align them according to the time stamp to ensure that the records of different systems can be analyzed in the same time window. Secondly, match the relevant data based on the same project number, order number, or equipment number to ensure that the information in the same business process can be integrated. In addition, for unstructured data in emails, documents, or business logs, natural language processing technology is used to calculate the text similarity to identify different data entries belonging to the same business process.

[0086] After the data fusion is completed, the present invention uses the association rule mining algorithm to calculate the support degree and confidence degree between departments.

[0087] After completing data fusion, the present invention uses an association rule mining algorithm to analyze the department collaboration pattern. Specifically, the present invention uses a frequent item set mining algorithm to extract stable department collaboration relationships from a large amount of business data. This method first constructs a transaction data set, where each transaction represents a complete business interaction, and all departments in the transaction form an item set. For example, in a software development enterprise, a complete product development process involves a product department, a research and development department, a testing department, and an operation department, and this business process can be represented as {product, research and development, testing, operation}. In a manufacturing enterprise, a complete production task involves research and development, production, quality management, and supply chain, and the transaction can be represented as {research and development, production, quality management, supply chain}. By constructing a large amount of business transaction data, the present invention can analyze the collaboration patterns of each department in different business scenarios through an association rule mining algorithm.

[0088] Specifically, the present invention first performs frequent item set mining, that is, counts the collaboration frequencies of each department within the enterprise and filters out department combinations with stable dependency relationships. The main steps of the algorithm are as follows:

[0089] Count the occurrence frequencies of each individual department, calculate the support, and filter out the departments with a support higher than the support threshold to form a frequent item set.

[0090] Based on the frequent item set, generate candidate department pairs, calculate the number of times they co-occur, and filter out the department pairs with a support higher than the support threshold.

[0091] Further expand to combinations of multiple departments, calculate more complex dependency relationships until no new high-frequency collaboration department combinations can be found.

[0092] Exemplarily, in the transaction data set of an enterprise, the following business records are included: {research and development, production, quality management}, {production, supply chain, procurement}, {research and development, production, supply chain}, and {production, supply chain}.

[0093] After scanning the data set, calculate the support of each department pair, that is, the frequency of co-occurrence of two departments in the same transaction. For example:

[0094] {research and development, production} co-occurs 3 times, with a total of 4 transactions, support = 3 / 4 = 0.75;

[0095] {production, supply chain} co-occurs 3 times, with a total of 4 transactions, support = 3 / 4 = 0.75;

[0096] {research and development, supply chain} co-occurs 1 time, with a total of 4 transactions, support = 1 / 4 = 0.25;

[0097] If the support threshold set by the enterprise is 0.5, then {R & D, Supply Chain} is screened out, while {R & D, Production} and {Production, Supply Chain} are retained.

[0098] Specifically, after obtaining the frequent item sets, the present invention calculates the confidence level to measure the strength of the dependence relationship between departments. The calculation method of the confidence level is as follows:

[0099] Confidence level = (the number of co-occurrences of Department A and Department B) / the number of occurrences of Department A

[0100] Exemplarily, the support from Production to Supply Chain is 0.75, and Production appears a total of 4 times. Therefore, the confidence level is 0.75; the support from R & D to Production is 0.75, and R & D appears a total of 3 times. Therefore, the confidence level is 1.0.

[0101] Specifically, the present invention sets a confidence level threshold, such as 0.6, that is, if when Department A occurs, there is at least a 60% probability that Department B will also occur, then it is determined that there is a stable dependence of A on B. For example, the confidence level from Production to Supply Chain is 0.75, which meets the threshold, while the confidence level from R & D to Supply Chain is 0.25, which is lower than 0.6 and is therefore screened out. After the calculated support and confidence levels are both higher than the set threshold, the present invention uses graph theory methods to establish an initial dependence relationship network structure diagram of the enterprise. In this network, nodes represent each department in the enterprise; the connecting lines represent the dependence relationships between departments, and the direction is from the dependent party to the party being depended on; the weight of the connecting line is comprehensively calculated from the support and confidence levels, and the calculation method is as follows:

[0102] Dependence relationship weight = 0.5 × support + 0.5 × confidence level

[0103] For example, if the support from Production to Supply Chain is 0.75 and the confidence level is 0.75, then the weight of this dependence relationship is 0.75.

[0104] Finally, through data fusion and association rule mining, the present invention can accurately construct an initial dependence relationship network diagram within the enterprise, clearly showing the information flow and resource allocation relationships between departments, and providing reliable data support for subsequent dynamic analysis, key node identification, and resource optimization.

[0105] It should be noted that in the implementation process of manufacturing enterprises, due to the stable production process and the long-term existence of cross-departmental collaboration relationships, the enterprise hopes to only screen out the dependency relationships with strong business associations to optimize production scheduling and supply chain management. Therefore, when setting the support threshold, the enterprise first counts the historical interaction data between all departments and finds that the average number of interactions between each pair of departments accounts for about 3% of the total interaction volume. Considering the high stability of business processes in the manufacturing industry, the enterprise decides to adopt a support threshold of 0.05, which is higher than the average value. That is, only the department pairs with an interaction frequency ratio exceeding 5% are recognized as having dependency relationships. For example, in the past year, the R & D department and the production department had a total of 700 collaborations, accounting for 7% of the total interaction volume, meeting the threshold requirements. However, the number of interactions between the R & D department and the procurement department was only 250 times, accounting for 2.5%, which is lower than the set threshold. Therefore, it is not considered that there is a stable dependency relationship between the two. In terms of setting the confidence threshold, the manufacturing enterprise focuses on the collaboration tightness of the core business processes and statistically finds that the average collaboration confidence between core departments is 0.6. Therefore, the enterprise sets the confidence threshold at 0.6, that is, when the behavior of a certain department depends on another department by at least 60%, it is recognized as a stable dependency. For example, the collaboration confidence between the production department and the supply chain department is 0.72, meeting the dependency relationship screening criteria, while the collaboration confidence between the R & D department and the procurement department is only 0.4, lower than the set threshold, so it is not included in the initial network diagram. Finally, the initial network diagram constructed by this manufacturing enterprise includes the stable dependency relationships among R & D, production, and supply chain, providing data support for subsequent resource optimization and risk assessment.

[0106] Exemplarily, in the implementation process of an Internet enterprise, due to the flexible business model and relatively dynamic collaboration methods among departments, the enterprise hopes to identify more potential dependency relationships to optimize product development and market operation strategies. Therefore, when setting the support threshold, the enterprise counts all the task assignment records among departments in the past year and finds that the interaction frequency between each pair of departments on average accounts for 2% of the total interaction volume. Considering the high collaboration frequency but relatively loose dependency relationships in Internet enterprises, the enterprise decides to adopt a relatively low support threshold of 0.02, that is, only department pairs with an interaction frequency ratio exceeding 2% are included in the dependency network. For example, the product department and the R & D department collaborate in 500 tasks, accounting for 10%, meeting the threshold requirements, while the interaction frequency between the marketing department and the R & D department is only 80 times, accounting for 1.6%, lower than the threshold, so it is not included in the initial network diagram. In terms of setting the confidence threshold, the collaboration mode of Internet enterprises is relatively flexible, and the average collaboration confidence among core departments is about 0.5. Therefore, the enterprise sets the confidence threshold at 0.5, that is, when the behavior of a certain department depends on another department by at least 50%, it is considered a stable dependency. For example, the collaboration confidence between the R & D department and the testing department is 0.7, meeting the dependency relationship screening criteria, while the collaboration confidence between the marketing department and the R & D department is only 0.3, lower than the threshold, so it is not included in the initial network diagram.

[0107] In step S13, it is necessary to perform clustering analysis on the connection strength and path length based on the initial network diagram, dynamically reconstruct the network, and fuse community mining and path optimization to identify key nodes, so as to obtain a dynamic network diagram of information flow among departments, including:

[0108] Extract the connection strength data and path length data from the initial network diagram;

[0109] Perform resource allocation mode analysis based on the connection strength data and the path length data to obtain the main transmission paths of department - to - department dependency relationships;

[0110] Perform dependency trend analysis on the main transmission paths to obtain the dependency trends among departments;

[0111] When the dependency trend is greater than the preset trend range, update the network structure of the initial network diagram based on graph theory methods to obtain an updated network diagram of department - to - department dependency relationships;

[0112] Identify key nodes for resource allocation based on the updated network diagram to obtain priority nodes that require priority resource allocation;

[0113] Perform path weight analysis on the priority nodes to obtain weight factors reflecting the importance of information transmission among departments;

[0114] Perform an operation to update the dependencies between departments according to the weight factor, and obtain a dynamic network diagram of information flow between departments.

[0115] First, extract the connection strength data and path length data from the initial network diagram. Among them, the connection strength represents the frequency of information exchange or the degree of resource sharing between different departments in the enterprise. For example, the number of email exchanges, document sharing times, and task handover quantities between two departments. The connection strength can be represented as a weight matrix , where represents the department to the department the interaction frequency between them, and the calculation method is as follows:

[0116]

[0117] For example, in a certain manufacturing enterprise, the R & D department sends 20 technical reports to the production department every week, and the total number of reports between all departments within the enterprise is 200. Then the connection strength between the R & D department and the production department is 0.1. The path length data is used to measure the minimum number of steps required for information to spread from one department to another, that is, the minimum number of hops of information in the enterprise network. The present invention uses the Dijkstra shortest path algorithm or the Floyd - Warshall algorithm to calculate the shortest path between departments to analyze the optimal transmission path of information flow within the enterprise. For example, in the enterprise network, the marketing department (M) needs to transmit customer feedback to the R & D department (R), but the marketing department cannot communicate directly with the R & D department and must transit through the product department (P). Then the path can be expressed as M→P→R, and the path length is 2, indicating that the information transmission requires two steps. If the marketing department can also reach the R & D department through the operations department (O), then the path M→O→R exists, and the system will compare the lengths of the two paths and select the shortest path. For example, in the Dijkstra algorithm, if the edge weight of M→P is 1, the edge weight of P→R is 1, the edge weight of M→O is 2, and the edge weight of O→R is 1, then the shortest path is still M→P→R, and the path length is 2, while the length of M→O→R is 3. Therefore, the marketing department should give priority to transmitting customer feedback to the R & D department through the product department to ensure the efficient flow of information. For more complex multi - department collaboration situations, the present invention uses the Floyd - Warshall algorithm to calculate the shortest path matrix between all departments to ensure the optimized transmission of information flow in the enterprise network and improve the efficiency of business collaboration.

[0118] Specifically, after obtaining the connection strength and path length, the present invention analyzes the resource allocation pattern of data based on a clustering algorithm to identify the main transmission paths between departments. Specifically, the K-Means clustering algorithm is used to classify similar departments to find groups of departments with close business interactions. For example, in a software company, the R & D department, the testing department, and the operation and maintenance department have a high connection strength and can be classified into the same business group, while the marketing, sales, and customer support departments can be classified into another group due to their frequent interactions. This clustering analysis helps to identify the key paths of information dissemination in the enterprise and ensures that resource allocation can prioritize high-frequency business needs.

[0119] The present invention further performs a dependency trend analysis on the identified main transmission paths based on a time series analysis algorithm to predict the change trend of the dependency relationship between departments. The time series analysis uses a sliding window method to calculate the change rate of department interactions over a past period of time :

[0120]

[0121] where represents the connection strength at the current time and represents the connection strength in the previous time period. For example, in a manufacturing enterprise, if the number of technical interaction reports between the R & D department and the production department in a certain quarter increases from 20 to 40, the dependency trend of this path is 1.0.

[0122] Specifically, if the dependency trend exceeds a preset trend range (for example, the set dependency growth threshold is 0.5), it indicates that the dependency degree of this path is increasing and the network structure needs to be updated.

[0123] During the network structure update process, the present invention dynamically adjusts the initial network diagram based on graph theory methods. When the connection strength of a certain department increases beyond the threshold or the path length shortens, the network structure will be automatically adjusted. For example, when the interaction frequency between the sales department and the product department increases by 200%, it means that the product improvement cycle is shortened, and the weight of this path needs to be increased to reflect the direct impact of market demand on R & D. After the network is updated, the present invention uses a community discovery algorithm to analyze the updated network structure to identify the key nodes that require priority resource allocation. The community discovery algorithm uses a modularity optimization method to group departments and calculate the belonging degree of departments in different communities. For example:

[0124]

[0125] where represents the proportion of edges within the community, It represents the proportion of the node degrees in the community. If the modularity of a certain department is high, it indicates that it plays a key role in the network. For example, in a manufacturing enterprise, the supply chain management department has dependencies with multiple departments and has a relatively high modularity, so it should be used as a node for priority resource allocation.

[0126] Specifically, after determining the priority nodes, the present invention uses a path optimization algorithm to perform path weight analysis on them and calculate the importance of information transmission. The path optimization uses the Analytic Hierarchy Process (AHP) to assign weights to different paths to ensure that key business paths are preferentially optimized. For example, in an enterprise, the product R & D path can be divided into market feedback → product planning → R & D design → production manufacturing. Among them, the influence weights of market feedback and product planning are relatively large, so resources should be preferentially allocated to make its process more efficient. Finally, the present invention combines the optimized weight factors and updates the enterprise network structure through graph theory methods to generate the final dynamic network diagram of information flow, ensuring that the network structure can reflect the information flow trend within the enterprise in real time and improving the scientificity and adaptability of resource management.

[0127] In step S14, it is necessary to extract the non-linear characteristics of the interaction of the dynamic network diagram to obtain the predictive analysis data for judging the node influence, including:

[0128] Extract the node connection relationship, weight data, and time series data in the dynamic network diagram;

[0129] Calculate the influence index of each node according to the node connection relationship and the weight data;

[0130] When the influence index is greater than the preset influence threshold, the node corresponding to the influence index is determined as the main node with an amplification effect;

[0131] When the influence index is less than the preset influence threshold, the node corresponding to the influence index is determined as the secondary node with a weakening effect;

[0132] Based on the time series data, perform an evolutionary trend analysis on the main nodes and secondary nodes to obtain the predictive analysis data of the node influence evolutionary trend.

[0133] First, it is necessary to extract node connection relationships, weight data, and time series data from the dynamic network diagram. The node connection relationship describes the information interaction situation among various departments within the enterprise and can be represented by establishing an adjacency matrix. For example, in a manufacturing enterprise, if the R & D department shares technical documents and conducts process collaboration with the production department, there is a direct information connection between the two. The weight data is used to measure the frequency and intensity of interaction between departments. For example, if the marketing department provides 100 customer feedbacks to the R & D department every month, while the total interaction volume with other departments is 500, the interaction weight of this path accounts for 20%, indicating that the marketing department has a greater impact on the R & D department. In addition, the time series data records the change of information flow intensity over time. For example, the order approval volume between the purchasing department and the supply chain management department shows an increasing or decreasing trend in different months, and the dynamic evolution of resource allocation can be analyzed based on this.

[0134] After the data extraction is completed, non - linear regression analysis is used to calculate the influence index of each department to measure its core role in the enterprise information network. The influence index is comprehensively calculated from multiple factors, including the number of directly connected departments (degree centrality), the role of serving as a transit in information transfer (betweenness centrality), and the interaction intensity (weight factor). For example, if a department has direct connections with multiple core business departments and plays a key bridging role in information transmission, its influence index is relatively high. On the contrary, if a department has a large number of connections but a low interaction frequency, or its information flow has little impact on other departments, the influence index is relatively low.

[0135] Specifically, the degree centrality represents the number of other nodes directly connected to a certain node, and the calculation method is as follows:

[0136]

[0137] Among them, is the connection value; is the target node for which the centrality is to be calculated; is to traverse all other nodes in the network except node outside.

[0138] For example, if a department is directly connected to 7 other departments, then indicates that this department has a strong direct influence.

[0139] The betweenness centrality represents the role of this node as a transit bridge in the network, that is, whether the internal information dissemination of the enterprise depends on this node. The calculation method is:

[0140]

[0141] Among them, is the total number of the shortest paths from the node to , and is the number of these paths passing through the node . For example, if a procurement department is involved in 15 out of 20 procurement approval processes, then .

[0142] Weight influence factor : Represents the interaction intensity of the node, and the calculation method is as follows:

[0143]

[0144] where is the weight value of the connection between node i and node j, quantifying the intensity or frequency of their interaction; is the target node for which the weight influence factor is to be calculated; is to traverse all adjacent nodes connected to node in the network.

[0145] For example, if the total interaction weight of a certain node is 0.6, it indicates that it has a greater impact on information flow. Finally, the node influence index can be calculated by a non - linear regression formula:

[0146]

[0147] where is the degree centrality; is the betweenness centrality; is the weight influence factor; is the time - series growth rate, characterizing the historical trend change and reflecting the dynamic development potential (such as the quarterly growth rate of the business scale of a department).

[0148] where is the regression weight, which can be obtained through training with historical data. Specifically, first collect data such as the degree centrality, betweenness centrality, weight factor, and time - series growth rate of each department in an enterprise over a past period of time, and label the actual business impact of each node. Subsequently, use a multiple regression model, with the influence index as the dependent variable and each centrality index as the independent variable, and use the least - squares method (OLS) or the gradient - descent method to train the regression model to determine the optimal weight coefficients. For example, in a manufacturing enterprise, by analyzing the data of the past year, it is found that the degree centrality contributes the most to the influence index, and after training the regression weight, , indicating that the enterprise's information flow mainly depends on the number of direct connections, followed by the relay role of information transmission, while the historical trend has relatively little impact. After training, the weight parameters are applied to new data to dynamically calculate the influence index of each department, and the enterprise's resource allocation and organizational structure are optimized accordingly.

[0149] To further distinguish key nodes from secondary nodes, the present invention sets an influence threshold to ensure accurate identification of departments in the enterprise network that play a key role in information flow and resource allocation. The influence index is obtained by statistical analysis of historical data and is classified using the method of mean ± standard deviation. Specifically, the main node threshold is set , where is the average influence of all departments, is the standard deviation, takes 1 or 1.5, adjusted according to the actual business. For example, in a manufacturing enterprise, the average influence is 0.5 and the standard deviation is 0.15. If k = 1, then , departments with an influence index higher than 0.65 (such as the R & D center and the core production department) are marked as main nodes, indicating that they undertake core decision-making or resource allocation functions in the enterprise operation. On the contrary, the secondary node threshold is set , if , then departments with an influence index lower than 0.35 (such as administrative logistics, human resources, and customer service centers) are marked as secondary nodes. These departments have less impact on the overall information flow and have a relatively lower priority in resource optimization. Departments between and (such as quality management and procurement departments) are classified as ordinary nodes, undertaking general operations and not belonging to the information flow hub.

[0150] In addition, this step combines time series analysis to predict the evolution trend of the influence of each department, in order to identify potential key nodes or weakening nodes in advance, so as to optimize the enterprise's resource allocation strategy. First, the system collects information flow data of each department over a past period of time, including core indicators such as business interaction frequency, task assignment situation, and number of approval processes, and constructs a time series data set for analyzing the dynamic changes of the influence of each department. To ensure the accuracy of the analysis, methods such as moving average method, exponential smoothing method, or long short-term memory (LSTM) neural network are used to model the time series data and extract the growth or decline trend of the influence index. Based on the trend analysis, departments that will become key nodes in the future can be identified, and the enterprise's management and resource allocation strategies can be adjusted in advance.

[0151] For example, in a manufacturing enterprise, the customer feedback data of the marketing department has shown an increasing trend in the past six months, from 120 to 300, indicating that the impact of market feedback on enterprise decision-making is increasing. By calculating the month-on-month growth rate, such as:

[0152]

[0153] Among them, the formula is used to calculate the month-on-month growth rate of the influence index of a certain department during a certain period of time, that is, to measure the change range of the influence of this department between two consecutive time periods (such as days, weeks, months). During the period, that is, to measure the change range of the influence of this department between two consecutive time periods (such as days, weeks, months). Represents the influence growth rate at time That is, the degree of growth or decline of the influence of this department within the current time period; Represents the time The influence index of this department at the moment, reflecting the importance of this department in enterprise information flow, decision-making participation and resource allocation; Represents the time The influence index of this department at the moment, that is, the influence value of the previous time period.

[0154] Exemplarily, if the growth rate of a certain department remains above 15% for multiple consecutive periods (such as the growth rates of the marketing department are 12.5%, 18.5% and 20% respectively), it can be considered that the influence of this department on enterprise operation will continue to increase in the future. Enterprises can adjust the resource allocation accordingly, such as increasing market data analysts, optimizing the R & D and marketing collaboration processes, or adjusting the production line to match the rapid changes in market demand, so as to improve the overall response speed and resource utilization rate of the enterprise.

[0155] On the contrary, if the influence index of a certain department continues to decline, such as the production department of a certain product line is gradually marginalized due to the reduction of market demand, the enterprise can use the same method for detection. For example, the order volume of a certain production department has continued to decline in the past six months, from 500 copies to 350 copies, and its month-on-month decline rates are -6.25% and -10.26% respectively. When this value is lower than the set lower limit threshold (such as -5%) for a long time, it indicates that the business of this department is in decline. At this time, the enterprise can consider adjusting the resource allocation of this department, such as optimizing the production line layout, reducing manpower and equipment investment, or even integrating or transforming this business unit to reduce resource waste and improve the overall operation efficiency.

[0156] In order to further improve the stability of prediction, the present invention uses the exponential smoothing method to reduce noise for short-term fluctuations and predict the future influence index. The exponential smoothing model is calculated as follows:

[0157]

[0158] Among them, Represents the influence index of this department at time The moment, reflecting the importance of this department in enterprise information flow, decision-making participation and resource allocation; is the predicted value for the previous period; is the predicted value for the future. is the smoothing factor (set to 0.8 - 0.9), which is used to balance the influence of historical data. For example, if the influence index of a production department in the current period is 350 and the index in the previous period is 390, then the predicted influence index for the next period is: = 356. If the predicted values for multiple future periods still show a downward trend, the enterprise can consider making early business adjustments, such as reducing relevant job allocations, reallocating production tasks, or seeking new business growth points to ensure the optimal utilization of overall resources.

[0159] In step S15, it is necessary to identify the nodes in the network that affect resource allocation and information transmission based on the dynamic network diagram and the predictive analysis data, and obtain key nodes and high-risk nodes, including:

[0160] Extract the node connection data, shortest path data, network topology change trend, clustering coefficient, and community membership data from the dynamic network diagram;

[0161] Based on the predictive analysis data, combined with the node connection data, the shortest path data, and the network topology change trend, perform an operation to correct the dynamic weight of the centrality index to obtain the centrality index for evaluating the importance of nodes in information dissemination;

[0162] Based on the predictive analysis data, combined with the clustering coefficient and the community membership data, perform an operation to correct the dynamic weight of the vulnerability index to obtain the vulnerability index for evaluating the dependence degree of nodes in information dissemination;

[0163] According to the centrality index, construct the eigenvectors of each node to obtain the centrality eigenvector;

[0164] Use the K-means clustering algorithm to perform importance ranking and screening on the centrality eigenvector to obtain preliminary key nodes with high importance, medium importance, and low importance;

[0165] Rank the importance of the preliminary key nodes to obtain the key nodes with high centrality;

[0166] According to the vulnerability index, analyze the connection characteristics and load capacity index of each node to obtain the vulnerability evaluation value;

[0167] When the vulnerability evaluation value is less than the preset vulnerability threshold, determine the node corresponding to the vulnerability evaluation value as a potential high-risk node;

[0168] Use the support vector machine algorithm to perform node failure prediction on the potential high-risk nodes to obtain the node failure probability;

[0169] Extract the set of nodes whose failure probability is greater than the preset probability threshold, and determine all nodes in the node set as high-risk nodes.

[0170] First, extract node connection data, shortest path data, network topology change trend, clustering coefficient, and community membership data from the dynamic network diagram. These data are used to construct the basic information of the network structure. Among them, the node connection data reflects the direct information interaction relationship between various departments, the shortest path data is used to calculate the optimal path of information flow, the network topology change trend monitors the dynamic adjustment of the network structure, the clustering coefficient is used to measure the closeness of a certain department to the surrounding departments, and the community membership data is used to analyze whether a department belongs to a certain stable organizational structure. The acquisition of these data depends on the dynamic update of the network model to ensure that the analyzed dependency relationships can accurately reflect the real-time state of the enterprise organization.

[0171] Next, based on the predictive analysis data, combined with the above-extracted data, correct the centrality indicators and dynamically adjust the weights of the centrality indicators to evaluate the importance of each node in information dissemination. Specifically, the centrality indicators include degree centrality, betweenness centrality, and closeness centrality. Among them, degree centrality measures the number of direct connections of a node, betweenness centrality measures the transit role of the node in the shortest path, and closeness centrality is used to evaluate the reachability of the node in the entire network. During the correction process, the present invention adopts a dynamic weight adjustment strategy to adjust the weights of different centrality indicators according to the predictive analysis data. For example, during the enterprise business adjustment period, the propagation mode of information flow will change, and the influence of certain key departments will increase. At this time, the weight of betweenness centrality can be increased to more accurately identify the nodes that play a pivotal role in the information transmission process.

[0172] At the same time, in order to identify high-risk nodes, it is necessary to calculate the vulnerability indicator, which is used to measure the degree of dependence of each node on information flow and resource allocation. The vulnerability indicator is mainly corrected by combining the clustering coefficient and community membership data to ensure that the risk analysis can reflect the actual business dependency relationship. Specifically, a node with a lower clustering coefficient means that its connection in the organizational structure is relatively isolated. Once this node has an abnormality, it will be more likely to cause information interruption or resource allocation failure. At the same time, a node with a lower community membership indicates that the department has a weaker relevance in the organization and its stability is relatively poor. Therefore, during the correction process, the system will adjust the weights of the clustering coefficient and community membership data to optimize the calculation of the vulnerability indicator.

[0173] After calculating the centrality and vulnerability metrics, the next step is to construct the centrality feature vectors for subsequent identification of key nodes. The feature vectors contain the degree centrality, betweenness centrality, closeness centrality of each node and their corrected weights, ensuring that the screening of key nodes in the network can accurately reflect their roles in resource allocation and information dissemination. Subsequently, the K-means clustering algorithm is used to perform a classification analysis on these feature vectors, dividing all nodes into three levels: high importance, medium importance, and low importance, thus forming a preliminary set of key nodes. This process can effectively distinguish the core business departments from the general business support departments, ensuring that the enterprise can prioritize the optimization of resource allocation for key departments.

[0174] After initially screening out the key nodes, the system will further rank the importance of these nodes to screen out the key nodes with high centrality. Specifically, calculate the influence index of the key nodes and sort them from high to low. For example, in an enterprise network, the R & D center, marketing department, and core production departments will be identified as high centrality nodes, while functional departments such as administrative logistics and financial audit are in the lower centrality range. Through this process, the enterprise can identify the core departments that truly affect resource flow and information transmission and adjust the resource investment strategy.

[0175] To evaluate potential high-risk nodes, the present invention further calculates the vulnerability assessment value and analyzes it in combination with the connection characteristics and load capacity of the nodes. When the vulnerability assessment value of a certain node is less than the preset vulnerability threshold, it indicates that its dependence on information flow and resource allocation is relatively high and its stability is relatively poor, then this node is marked as a potential high-risk node. For example, in a supply chain management system, if a certain supplier node has a small number of connections in the network (low degree centrality) and a low belonging degree to the supply chain community it belongs to, it means that this node lacks alternative supply channels and will affect the entire production process if an abnormality occurs, so it should be marked as a high risk.

[0176] After determining the potential high-risk nodes, the present invention uses the support vector machine (SVM) algorithm to predict node failures and calculates the node failure probability of each potential high-risk node. The support vector machine analyzes the dynamic feature evolution law of high-risk nodes in historical data to predict the failure probability of a specific node within a future time window. For example, in a financial system, if the fund flow record of a certain business department shows abnormal fluctuations, indicating the risk of its future capital chain breakage, the system can train the support vector machine model based on historical transaction data to give a risk warning for this department. When the failure probability of a certain node is greater than the preset probability threshold (such as 70%), then this node will be finally marked as a high-risk node, and the enterprise can take measures in advance, such as optimizing its information transmission path, adjusting resource allocation, or establishing alternative solutions to reduce potential losses.

[0177] The following uses a specific application scenario to illustrate step S15 of the present invention:

[0178] In the supply chain management of a certain intelligent manufacturing enterprise, the information flow and resource allocation relationships among different departments are complex. Especially in the links of raw material supply, production, warehousing, sales, etc., the collaborative efficiency of each department directly affects the operational stability of the enterprise. However, due to the dynamic and non-linear dependence relationships of the supply chain, it is difficult for enterprises to accurately identify key nodes and high-risk nodes, resulting in unreasonable resource allocation and even the risk of supply chain breakage. In order to optimize supply chain management, the present invention uses dynamic network graph analysis and prediction models to identify key nodes and high-risk nodes in the enterprise's supply chain, optimize resource allocation strategies, and improve operational stability and resilience.

[0179] First, extract supply chain network data from the enterprise's supply chain management system (SCM), production execution system (MES), and enterprise resource planning system (ERP). This data includes information flow and resource allocation in multiple dimensions. For example, during the production process, the raw material procurement department needs to maintain stable information interaction with the production department to ensure the timely supply of raw materials; the production department shares finished product inventory data with the warehousing department to reasonably arrange production plans and inventory management. To quantify these dependence relationships, the system extracts node connection data, shortest path data, network topology change trends, clustering coefficients, and community membership data. Among them, node connection data is used to record the direct interaction relationships between departments, such as the cooperation network between the procurement department and multiple suppliers; shortest path data is used to calculate the optimal path for information circulation and logistics allocation, such as the transportation route of raw materials from suppliers to the production line; network topology change trends are used to monitor the structural adjustment of the supply chain, such as the addition of new suppliers or production line changes; the clustering coefficient measures the tightness of a certain department in the supply chain. For example, if the warehousing department is closely connected to multiple production lines, its role in resource allocation is relatively large; community membership data is used to divide business modules in the supply chain. For example, the raw material procurement, production, and quality inspection departments are classified as the "production community", and the sales, logistics, and customer service departments are classified as the "sales community" to identify the collaborative relationships of each link in the supply chain.

[0180] After data acquisition, the system calculates centrality metrics to identify critical nodes in the supply chain. Degree centrality measures the number of direct connections of a certain department in the supply chain. For example, if the procurement department has established long-term cooperative relationships with 10 suppliers, its degree centrality is relatively high, indicating that this department is crucial in the raw material supply chain. Betweenness centrality measures the transit role of a certain node in the supply chain network. For example, if the warehousing department undertakes multiple functions such as finished product allocation and raw material storage, then the betweenness centrality of this node is relatively high, indicating that it plays a key role in information and resource transfer. Closeness centrality is used to evaluate the accessibility of a certain department in the entire supply chain. For example, if the production department has a high closeness centrality, it means that it has direct interactions with multiple business modules, indicating that this department is crucial for the operation of the entire supply chain. By calculating these metrics and combining them with predictive analysis data, the system can dynamically adjust the weights of each metric to ensure the accuracy of identifying critical nodes.

[0181] After identifying critical nodes, the system further calculates vulnerability metrics to identify high-risk nodes in the supply chain. First, by combining clustering coefficient and community membership data, the stability of a certain node is evaluated. If a certain supplier has a low clustering coefficient, it means that the cooperation between this supplier and the enterprise is relatively independent, and once there are supply chain problems, the impact will be greater. For example, if enterprise A only relies on a single supplier B to provide key raw materials, and if supplier B experiences a supply chain break, then the production capacity of enterprise A will drop significantly. Therefore, the vulnerability assessment value of supplier B is relatively high and it needs to be marked as a high-risk node. In addition, if a certain node has a low community membership, it indicates that the degree of collaboration of this department in the supply chain network is not high. For example, if a logistics company only has temporary cooperative relationships with enterprises and does not have a fixed transportation network support, once this logistics company stops providing services, the delivery chain of the enterprise will be interrupted, so its vulnerability is relatively high. Similarly, if a production line node has a small number of connections but undertakes the production task of key products, then the supply chain dependence of this node is relatively high and its risk is also relatively large.

[0182] After determining potential high-risk nodes, the system uses the Support Vector Machine (SVM) algorithm to predict node failures and calculates the node failure probability of each potential high-risk node. The support vector machine predicts the failure probability of a specific node within a future time window by analyzing the evolution law of the dynamic characteristics of high-risk nodes in historical data. For example, if the historical transaction data of a certain supplier shows that its supply cycle has fluctuated greatly in the past three months, the inventory turnover rate is low, and the capital flow is restricted, then its supply stability is poor and the failure probability exceeds 70%. At this time, the system automatically marks it as a high-risk node and issues a warning to enterprise managers, suggesting adjusting the supply chain strategy, such as finding alternative suppliers, increasing inventory reserves, or optimizing the procurement cycle to reduce supply chain risks.

[0183] In step S16, it is necessary to identify the critical propagation paths in the network based on the critical nodes and the high-risk nodes, and obtain the critical propagation paths that affect resource allocation and information transmission, including:

[0184] Construct a risk network graph using graph theory methods based on the critical nodes and the high-risk nodes;

[0185] Perform path parameter calculations on the risk network graph to obtain the propagation data and stability data of the nodes on the path;

[0186] Evaluate the node importance and path stability based on the propagation data and stability data to obtain a path evaluation result;

[0187] Group the path evaluation results according to the path propagation priority, resource influence, and decision chain weight to obtain the critical propagation paths that affect resource allocation and information transmission.

[0188] First, the present invention uses graph theory methods to construct a risk network graph to represent the information flow paths and their potential risks within an enterprise. In the risk network graph, critical nodes are the cores of information flow and resource allocation, such as the R & D department, the marketing department, core suppliers, etc. These nodes have high degree centrality, betweenness centrality, or closeness centrality, and play a pivotal role in organizational decision-making and operation management. At the same time, high-risk nodes are the nodes that affect information circulation and resource supply due to insufficient resource carrying capacity, high dependence, or high failure probability, such as a single supplier, a critical equipment maintenance node, or a department with greater financial pressure. To quantify these dependencies, the system constructs an adjacency matrix A to represent the information flow and resource flow relationships between departments, and the values in the matrix represent the interaction intensity between departments. For example, in a certain manufacturing enterprise, the R & D department has more information interactions with the production department, so its edge weight is higher, while the direct interaction between the finance department and the production department is less, so its edge weight is lower. Through this risk network graph, the paths of information flow and resource flow can be clearly observed, and their propagation characteristics can be further analyzed.

[0189] After constructing the risk network graph, the present invention uses the shortest path algorithm (such as the Dijkstra algorithm or the Floyd-Warshall algorithm) to calculate the path propagation data and stability data in the network. Path propagation measures the transmission efficiency of information or resources on this path, while path stability measures whether this path is vulnerable to the influence of high-risk nodes. Specifically, the system calculates all information transmission paths and uses the following formula to calculate path propagation:

[0190]

[0191] where Represents the betweenness centrality of each node on the path. The higher this value, the more important the node is in information transmission and the greater the path propagation ability. The greater it is, the greater the influence of the path on resource allocation and decision-making. At the same time, considering the influence of high-risk nodes, calculate the stability weight on the path:

[0192]

[0193] Where, Represents the failure risk of each node on the path. The higher this value, the lower the stability of the path. Therefore, the higher the path stability is, the more stable the path is.

[0194] After calculating the path propagation ability and stability, the present invention uses a weighted algorithm to comprehensively evaluate the importance of the path. The importance of the path is calculated by the following formula:

[0195]

[0196] Where, , is the weight coefficient. The adjustment of the weight coefficient needs to be optimized according to the management requirements and risk preferences of the enterprise to ensure that the evaluation of key propagation paths is more in line with the actual business scenario. When the enterprise pays more attention to the efficiency of information transmission, the propagation weight can be appropriately increased, so that the paths with smooth information flow obtain higher priority in the optimization process; on the contrary, when the enterprise pays more attention to system stability and risk prevention, the stability weight can be increased, so that the paths with higher stability and less influence from high-risk nodes are more emphasized. In addition, the enterprise can optimize the weight setting through historical data analysis and backtesting. For example, calculate whether there are often information lag or resource allocation problems in the paths with high propagation but low stability in the past period. If so, it is necessary to appropriately increase ; if some paths with high propagation ability can effectively support resource allocation in the past decision-making process, can be increased.

[0197] After determining the importance of the paths, the present invention further classifies the path evaluation results using a clustering algorithm (such as K-means or hierarchical clustering) to better understand and optimize the management of the enterprise's information flow and resource flow. The key propagation paths are mainly classified into the following categories: high-priority paths, that is, paths with high and stable propagation efficiency, such as R & D → production → logistics, which are crucial for the transfer of enterprise resources and need to ensure their smoothness as a priority; high-risk paths, that is, paths with high propagation efficiency but low stability, such as key suppliers → procurement → production. If problems occur in this path, it will cause the enterprise's operation to be blocked. Therefore, it is necessary to focus on monitoring and formulating emergency plans; redundant paths, that is, paths with low propagation efficiency and high stability, such as daily management processes like administration → finance → audit, which can be used for general management optimization without excessive attention.

[0198] In step S17, it is necessary to predict the critical positions where resource bottlenecks and resource conflicts occur based on the key nodes, the high-risk nodes, and the key propagation paths, and generate resource optimization suggestions, including:

[0199] Analyze the resource allocation status according to the key nodes, the high-risk nodes, and the key propagation paths to obtain the resource usage of the key nodes, the resource pressure of the high-risk nodes, and the traffic load of the key propagation paths;

[0200] Predict the critical positions where resource bottlenecks and resource conflicts occur based on the resource usage, the resource pressure, and the traffic load, in combination with the preset historical resource scheduling data;

[0201] Adjust the resource allocation strategy for the critical positions to generate resource optimization suggestions.

[0202] First, the system extracts the key nodes, high-risk nodes, and key propagation path information obtained from the previous steps and analyzes their resource allocation status. Among them, the resource usage of key nodes includes the current resource occupancy rate, consumption rate, and remaining available resources of the node. For example, in a manufacturing enterprise, as a key node, the resource usage of the production department can be measured by indicators such as the operating rate of production equipment, the consumption rate of inventory raw materials, and the personnel configuration load. The resource pressure of high-risk nodes involves the stability of the node in resource scheduling. For example, a sole supplier, a single logistics channel, or a key server becomes the bottleneck of the system. This part of the analysis mainly calculates the resource redundancy (current inventory / demand), load ratio (current task volume / maximum carrying capacity), and historical resource fluctuation of the node to judge the stability of the node. The traffic load of the key propagation path measures the flow of resources or information on different paths, including indicators such as the resource transmission rate, average response time, and peak traffic of the path. For example, in a logistics transportation network, the transportation capacity from the supplier to the warehouse and from the warehouse to the production line can be evaluated by the hourly freight volume and historical delay conditions.

[0203] Specifically, after completing the resource status analysis, this step combines historical resource scheduling data and predicts the occurrence locations of resource bottlenecks and resource conflicts through machine learning algorithms. First, the regression analysis method is used to predict the resource demand trend of key nodes. Specifically, time series models (such as ARIMA, LSTM) are used to analyze the resource consumption data in the past period to predict the growth trend of future resource demand. For example, if the sales data shows that the market demand for a certain product is expected to increase by 30% in the next month, the system can predict that the material demand of the production department will increase synchronously during the corresponding period and adjust the procurement plan in advance. Second, classification algorithms (such as random forest, support vector machine SVM) are used to train the resource shortage risk model. By analyzing factors such as supply shortages and equipment failures in historical data, it is predicted which high-risk nodes will become bottlenecks due to resource shortages. For example, if a key supplier has experienced delivery delays in 4 months out of the past 12 months, the system can predict that there is still a 30%-40% risk of supply instability in the future and recommend adding backup suppliers. In addition, a traffic analysis model (such as Bayesian network) is used to predict the traffic changes on the key propagation path to identify potential path congestion problems in advance. For example, in an enterprise's information system, if it is predicted that the server access volume will surge by 50% during the annual audit period, the system can warn that the computing resource demand of this path is insufficient and recommend expanding the server capacity in advance.

[0204] Specifically, based on the prediction results of resource bottlenecks and conflict risks, the present invention adopts an optimization algorithm to adjust the resource allocation at key positions, so as to reduce systemic risks and improve resource utilization. The optimization methods include dynamic resource scheduling, resource redundancy optimization, and intelligent path adjustment. In terms of dynamic resource scheduling, the system redistributes available resources among different departments according to the resource demand prediction results. For example, when a production line faces the risk of shutdown due to a shortage of raw materials, the system can recommend transferring raw materials from other production lines with higher inventory levels to ensure continuous production. For resource redundancy optimization, if the system detects that the resource carrying capacity of a high-risk node is approaching the upper limit, such as the supply capacity of a key supplier tending to be saturated, the system can recommend adding backup resources, such as adding new suppliers or increasing the safety inventory level, to reduce supply chain risks. In terms of intelligent path adjustment, for the key propagation paths where resource circulation is blocked, the system can allocate resources by optimizing the path. For example, in a data transmission network, if the system predicts that the load of a certain server exceeds 80% during peak hours, some computing tasks can be transferred to the backup server in advance to improve the overall network stability.

[0205] In a specific embodiment, assume that a multinational manufacturing enterprise hopes to optimize the resource scheduling of its global supply chain. The system first analyzes the key nodes (production workshops, suppliers), high-risk nodes (single raw material supplier), and key propagation paths (supplier → warehouse → production line), and finds that the order fulfillment ability of a major supplier fluctuates, and the inventory level of the production line is low. Combining historical data analysis, the system predicts that there is a 10%-15% risk of supply delay for this supplier in the next three weeks, which will lead to production bottlenecks. Therefore, the system generates the following optimization suggestions: add backup suppliers to reduce the enterprise's dependence on a single supplier to ensure stable raw material supply; adjust the inventory management strategy, increase the inventory safety threshold, and increase the reserve of key materials to reduce the impact of supply chain fluctuations; optimize the production plan, by adjusting the production schedule, give priority to producing products that do not depend on the raw materials of this supplier to avoid shutdown due to out-of-stock; optimize the logistics path, if it is found that a certain transportation route affects the supply chain efficiency due to traffic congestion during peak hours, it can recommend adjusting the logistics route or increasing the number of transportation batches to ensure the timely delivery of raw materials.

[0206] In summary, the present invention provides an enterprise digital management method and system based on data mining. The present invention can identify the dynamic changes of implicit dependence relationships in enterprises, locate high-risk nodes and key propagation paths, and optimize resource allocation.

[0207] Referring to Figure 2 , the second embodiment of the present invention provides an enterprise digital management system based on data mining, including:

[0208] A data acquisition module for acquiring information flow data and resource allocation data between departments;

[0209] A dependency analysis module for performing dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of the inter-departmental dependency relationship;

[0210] A path optimization module for performing clustering analysis on connection strength and path length based on the initial network diagram, dynamically reconstructing the network, and integrating community mining and path optimization to identify key nodes, thereby obtaining a dynamic network diagram of information flow between departments;

[0211] A prediction analysis module for extracting non-linear features of interaction effects from the dynamic network diagram to obtain prediction analysis data for judging node influence;

[0212] A node identification module for identifying nodes in the network that affect resource allocation and information transmission based on the dynamic network diagram and the prediction analysis data to obtain key nodes and high-risk nodes;

[0213] A critical path module for identifying critical propagation paths in the network based on the key nodes and the high-risk nodes to obtain critical propagation paths that affect resource allocation and information transmission;

[0214] A resource optimization module for predicting critical locations where resource bottlenecks and resource conflicts occur based on the key nodes, the high-risk nodes, and the critical propagation paths, and generating resource optimization suggestions.

[0215] It should be noted that an enterprise digital management system based on data mining provided by an embodiment of the present invention is used to execute all process steps of a method for enterprise digital management based on data mining in the above embodiment. The working principles and beneficial effects of the two correspond one by one, and thus will not be elaborated here.

[0216] An embodiment of the present invention also provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a prediction analysis program. When the processor executes the computer program, the steps in each embodiment of the above-mentioned method for enterprise digital management based on data mining are implemented, such as Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in each device embodiment above are implemented, such as the node identification module.

[0217] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the electronic device.

[0218] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0219] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and lines.

[0220] The memory can be used to store the computer program and / or module. By running or executing the computer program and / or module stored in the memory, and invoking the data stored in the memory, the processor can implement various functions of the electronic device. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store the operating system, application programs required for at least one function (such as the sound playback function, image playback function, etc.); the data storage area can store the data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disks, memory, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0221] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0222] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0223] The specific embodiments described above have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An enterprise digital management method based on data mining, characterized in that Including: Obtaining information flow data and resource allocation data among departments; Performing dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of inter-departmental dependencies; Performing clustering analysis on connection strength and path length according to the initial network diagram, dynamically reconstructing the network, and integrating community mining and path optimization to identify key nodes, thereby obtaining a dynamic network diagram of information flow among departments; Performing extraction of non-linear characteristics of interaction effects on the dynamic network diagram to obtain predictive analysis data for judging node influence; Identifying nodes in the network that affect resource allocation and information transmission based on the dynamic network diagram and the predictive analysis data to obtain key nodes and high-risk nodes; Identifying key propagation paths in the network based on the key nodes and the high-risk nodes to obtain key propagation paths that affect resource allocation and information transmission; Predicting the key positions where resource bottlenecks and resource conflicts occur based on the key nodes, the high-risk nodes, and the key propagation paths, and generating resource optimization suggestions.

2. The enterprise digital management method based on data mining according to claim 1, characterized in that The step of performing dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of inter-departmental dependencies includes: Calculating the support degree and confidence degree among departments according to the information flow data and the resource allocation data; When the support degree and the confidence degree are respectively greater than a preset support degree threshold and a preset confidence degree threshold, establishing a network structure diagram of inter-departmental dependencies based on graph theory methods to obtain an initial network diagram of inter-departmental dependencies.

3. The enterprise digital management method based on data mining according to claim 1, characterized in that The step of performing clustering analysis on connection strength and path length according to the initial network diagram, dynamically reconstructing the network, and integrating community mining and path optimization to identify key nodes, thereby obtaining a dynamic network diagram of information flow among departments includes: Extracting connection strength data and path length data from the initial network diagram; Performing resource allocation mode analysis based on the connection strength data and the path length data to obtain the main transmission paths of inter-departmental dependencies; Performing dependency trend analysis on the main transmission paths to obtain the dependency trend among departments; When the dependency trend is greater than a preset trend range, updating the network structure of the initial network diagram based on graph theory methods to obtain an updated network diagram of inter-departmental dependencies; Identifying key nodes for resource allocation according to the updated network diagram to obtain priority nodes that require priority resource allocation; Performing path weight analysis on the priority nodes to obtain weight factors reflecting the importance of information transmission among departments; Performing an operation of updating inter-departmental dependencies according to the weight factors to obtain a dynamic network diagram of information flow among departments.

4. The enterprise digital management method based on data mining according to claim 1, wherein The step of performing extraction of non-linear characteristics of interaction effects on the dynamic network diagram to obtain predictive analysis data for judging node influence includes: Extracting node connection relationships, weight data, and time series data from the dynamic network diagram; Calculating the influence index of each node according to the node connection relationships and the weight data; When the influence index is greater than a preset influence threshold, determining the node corresponding to the influence index as a main node with an amplification effect; When the influence index is less than the preset influence threshold, the node corresponding to the influence index is determined as a secondary node with a weakening effect; Based on the time series data, perform an evolutionary trend analysis on the primary nodes and secondary nodes to obtain predictive analysis data on the evolutionary trend of node influence.

5. The enterprise digital management method based on data mining according to claim 1, wherein According to the dynamic network graph and the predictive analysis data, identify the nodes in the network that affect resource allocation and information transmission, and obtain key nodes and high-risk nodes, including: Extract the node connection data, shortest path data, network topology change trend, clustering coefficient, and community membership data in the dynamic network graph; Based on the predictive analysis data, combined with the node connection data, the shortest path data, and the network topology change trend, perform an operation on the dynamic weight of the modified centrality index to obtain a centrality index for evaluating the importance of nodes in information dissemination; Based on the predictive analysis data, combined with the clustering coefficient and the community membership data, perform a dynamic weight modification on the vulnerability index to obtain a vulnerability index for evaluating the dependence degree of nodes in information dissemination; According to the centrality index, construct feature vectors for each node to obtain a centrality feature vector; Use the K-means clustering algorithm to perform importance ranking and screening on the centrality feature vector to obtain preliminary key nodes with high importance, medium importance, and low importance; Rank the importance of the preliminary key nodes to obtain key nodes with high centrality; According to the vulnerability index, analyze the connection characteristics and load capacity index of each node to obtain a vulnerability evaluation value; When the vulnerability evaluation value is less than the preset vulnerability threshold, the node corresponding to the vulnerability evaluation value is determined as a potential high-risk node; Use the support vector machine algorithm to predict node failure for the potential high-risk nodes to obtain the node failure probability; Extract the set of nodes whose node failure probability is greater than the preset probability threshold, and determine all nodes in the node set as high-risk nodes.

6. The enterprise digital management method based on data mining according to claim 1, wherein According to the key nodes and the high-risk nodes, identify the key propagation paths in the network to obtain the key propagation paths that affect resource allocation and information transmission, including: According to the key nodes and the high-risk nodes, use graph theory methods to construct a risk network graph; Perform path parameter calculations on the risk network graph to obtain the propagation data and stability data of the nodes on the path; According to the propagation data and stability data, evaluate the node importance and path stability to obtain a path evaluation result; Group the path evaluation results according to the path propagation priority, resource influence, and decision chain weight to obtain the key propagation paths that affect resource allocation and information transmission.

7. The enterprise digital management method based on data mining according to claim 1, characterized in that Based on the key nodes, the high-risk nodes, and the key propagation paths, predict the critical locations where resource bottlenecks and resource conflicts occur, and generate resource optimization suggestions, including: According to the key nodes, the high-risk nodes, and the key propagation paths, analyze the resource allocation status to obtain the resource usage of key nodes, the resource pressure of high-risk nodes, and the traffic load of key propagation paths; Predict the critical locations where resource bottlenecks and resource conflicts occur based on the resource usage, resource pressure, and traffic load, in combination with the preset historical resource scheduling data; Adjust the resource allocation strategy for the critical locations to generate resource optimization suggestions.

8. An enterprise digital management system based on data mining, characterized in that, It includes: A data acquisition module for acquiring information flow data and resource allocation data between departments; A dependency analysis module for performing dependency relationship analysis based on the information flow data and the resource allocation data to obtain an initial network diagram of the inter-departmental dependency relationship; A path optimization module for performing clustering analysis of connection strength and path length based on the initial network diagram, dynamically reconstructing the network, and integrating community mining and path optimization to identify critical nodes, thereby obtaining a dynamic network diagram of the inter-departmental information flow; A prediction analysis module for extracting non-linear features of interaction effects from the dynamic network diagram to obtain prediction analysis data for judging node influence; A node identification module for identifying nodes in the network that affect resource allocation and information transmission based on the dynamic network diagram and the prediction analysis data to obtain critical nodes and high-risk nodes; A critical path module for identifying critical propagation paths in the network based on the critical nodes and the high-risk nodes to obtain critical propagation paths that affect resource allocation and information transmission; A resource optimization module for predicting the critical locations where resource bottlenecks and resource conflicts occur and generating resource optimization suggestions based on the critical nodes, the high-risk nodes, and the critical propagation paths; 9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the data mining-based enterprise digital management method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the data mining-based enterprise digital management method according to any one of claims 1 to 7.

Images