Data processing system and method in qualification authentication and management consultation field

Abstract

The invention provides a data processing system and method in the field of qualification authentication and management consultation. In the system, an interaction module receives a data collection request for qualification authentication or management consultation corresponding to a request input instruction; the data collection request comprises a data type and a to-be-collected data identifier; the data type comprises a qualification authentication type or a management consultation type; the data acquisition module determines a target domain knowledge graph corresponding to the data type by using stored domain knowledge graphs corresponding to different data types, determines an entity node related to the data identifier in the target domain knowledge graph, determines a to-be-acquired data source based on the entity node, and transmits the to-be-acquired data source to the data acquisition module; collecting target data corresponding to the entity node from the to-be-collected data source; and the data analysis module analyzes the target data based on the constructed data analysis model to obtain a data analysis result. According to the system, automation of data collection is realized, the accuracy and reliability of data are improved, and the risk of misjudgment is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of qualification certification and management consulting, and in particular, to a data processing system and method in the field of qualification certification and management consulting. Background Art

[0002] Enterprise qualification certification and management consulting is an interdisciplinary professional service field, covering legal compliance, quality management, strategic planning, operational optimization, etc. The core of this field is to help enterprises meet external regulatory requirements (qualification certification) and internal management needs (management consulting) to achieve sustainable development.

[0003] However, in the field of qualification certification and management consulting, enterprises need to collect a large amount of data related to qualification certification or management consulting. These data may involve the company's operating conditions, financial information, industry standards, etc. The current data collection method mainly relies on manual operations or simple automation tools, which is inefficient and prone to errors. The data sources are scattered and lack unified standards and structured processing methods. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a data processing system and method in the field of qualification certification and management consulting, which solves the above-mentioned problems existing in the prior art, realizes the automation of data collection, and at the same time utilizes the domain knowledge graphs corresponding to different data types stored in the database to ensure that the data collection and analysis process is based on an authoritative and professional knowledge system, thereby improving the accuracy and reliability of the data and reducing the risk of misjudgment.

[0005] In a first aspect, a data processing system in the field of qualification certification and management consulting is provided, which is applied to a data management network including a data management platform and multiple network partitions, each network partition is a sub-management network of an industry type to which a configured management enterprise belongs, each network partition includes multiple data processing systems and databases with the same structure, after receiving a first data processing request including an industry type and an enterprise identifier sent by a target enterprise, the data management platform verifies whether the target enterprise is a legal enterprise by inputting the enterprise identifier into a configured enterprise credit information system, if not, an instruction to refuse to process the first data processing request is generated; if yes, a second data processing request including the enterprise identifier is sent to the target network partition corresponding to the industry type; the enterprise identifier is a unified social credit code or an enterprise name; the target network partition uses a response time scheduling rule to schedule each data processing system to determine the data processing system that processes the second data processing request, the response time scheduling rule is a rule for selecting the data processing system with the shortest response time, the system includes: each data processing system includes an interaction module, a data acquisition module and a data analysis module; wherein: An interaction module is used to send a request input instruction to a user of a target enterprise based on the second data processing request, and receive a data collection request for qualification certification or management consulting corresponding to the request input instruction; the data collection request includes a data type and a data identifier to be collected; the data type includes a qualification certification type or a management consulting type; the data identifier is a data name related to the data type; A data collection module is used to determine the target domain knowledge graph corresponding to the data type by using the domain knowledge graphs corresponding to different data types stored in the database, and determine the entity node related to the data identifier in the target domain knowledge graph, determine the data source to be collected based on the entity node, and collect the target data corresponding to the entity node from the data source to be collected; The data analysis module is used to analyze the target data based on the constructed data analysis model to obtain data analysis results; the data analysis results include qualification assessment information or management diagnosis information.

[0006] In an optional implementation, the domain knowledge graphs corresponding to the different data types include a first domain knowledge graph corresponding to the qualification certification field and a second domain knowledge graph corresponding to the management consulting field; wherein: The first domain knowledge graph is constructed with the target enterprise's industry attribute entities, personnel entities, equipment and facility entities, qualification standard entities, and engineering project entities as nodes, and the relationships between entities of different types as edges; The second domain knowledge graph is constructed with the consulting service entities, department entities, management process entities, management tool entities and management problem entities of the target enterprise as nodes and the relationships between different types of entities as edges.

[0007] In an optional implementation, when there are multiple entity nodes related to the data identifier determined in the target domain knowledge graph, the data acquisition module is used to determine the initial entity node corresponding to the data identifier in the target domain knowledge graph; Count the first entity nodes that have one relationship edge with the initial entity node, the second entity nodes that have two relationship edges with the initial entity node, and the third entity nodes that have three relationship edges with the initial entity node; The initial entity node, the first entity node, the second entity node and the third entity node are determined as entity nodes associated with the data identifier.

[0008] In an optional implementation, the data analysis module is specifically used to: configure the data priority corresponding to the entity node based on the number of relationship edges between the entity nodes related to the data identifier; based on the constructed data analysis model, analyze the target data corresponding to each entity node according to the data priority to obtain the data analysis result.

[0009] In an optional implementation, the interaction module is further used to receive feedback information from different users through the user interface of the target enterprise, wherein the feedback information includes questions and / or improvement suggestions raised by the different users on the displayed data analysis results; The data analysis module is also used to classify the feedback information received from different users using a preset classification algorithm to obtain at least one feedback information cluster; the feedback information in each feedback information cluster is sequentially input into a trained key information extraction model to obtain the key information corresponding to each feedback information cluster, wherein the key information extraction model is obtained by training a classification model to be trained based on a feedback fragment training data set containing feedback fragments marked as key information and non-key information; based on the key information extracted from each feedback information cluster and the association relationship between different key information, an information knowledge graph representing the key information is constructed; the information knowledge graph uses the key concept entities corresponding to the key information as nodes, and the association relationship between the key information as edges between different key concept entities.

[0010] In an optional implementation, the data analysis module specifically uses a preset classification algorithm to classify the feedback information received from different users to obtain at least one feedback information cluster, including: For the field of qualification certification or management consulting, determine a first feature extraction module of historical feedback information and a second feature extraction module of historical feedback information, wherein the first feature extraction module and the second feature extraction module have the same structure but different parameters, and the first feature extraction module and the second feature extraction module are obtained by training based on historical feedback information; Inputting the feedback information in the current cycle into the first feature extraction module to obtain a first dense feature vector set of each feedback information, and inputting the first dense feature vector set into a predetermined similarity scoring module to obtain a first similarity score; Inputting the feedback information in the current cycle into the second feature extraction module to obtain a second dense feature vector set of each feedback information, and inputting the second dense feature vector set into a predetermined similarity scoring module to obtain a second similarity score; Merging the first similarity score and the second similarity score of each feedback information to obtain a fused similarity score of each feedback information; Based on the fusion similarity score, determining a first feedback information set belonging to the field of qualification certification and a second feedback information set belonging to the field of management consulting in the feedback information to be classified in the current cycle; A first sample is generated for each set of the first feedback information, and a corresponding second sample is generated for each set of the first feedback information using a graph encoder; a third sample is generated for each set of the second feedback information, and a corresponding fourth sample is generated for each set of the second feedback information using a graph encoder; the second sample corresponding to the same first feedback information is a positive sample of the first sample, the second sample, the third sample and the fourth sample corresponding to different first feedback information are negative samples of the first sample, the fourth sample corresponding to the same second feedback information is a positive sample of the third sample, and the fourth sample, the first sample and the second sample corresponding to different second feedback information are negative samples of the first sample; the parameters in the first feature extraction module are optimized by narrowing the difference between the first sample or the third sample and the positive sample, and by widening the difference between the first sample or the third sample and the negative sample, and the parameters of the second feature extraction module are optimized based on the historical parameters of the first feature extraction module, so as to obtain the first feature extraction module of the current period and the second feature extraction module of the current period; the feedback information in the current period is classified based on the first feature extraction module and the classifier of the current period to obtain at least one feedback information cluster.

[0011] In an optional implementation, the fusion similarity score is used to evaluate the similarity between the feedback information to be classified in the current cycle and the feedback information in the field of qualification certification or the feedback information in the field of management consulting.

[0012] In an optional implementation, the system also includes a visualization display module for displaying the data analysis results, and using preset display graphics to display the frequency distribution of key information extracted from different feedback information clusters, and using preset network graphics to display the information knowledge graph.

[0013] In an optional implementation, the constructed data analysis model includes a qualification certification analysis model and a management consulting analysis model; The construction process of the qualification certification analysis model includes: determining the historical certification target of the enterprise qualification certification and the key factors affecting the historical certification target; obtaining the comparison results of the experts comparing different key factors in pairs, and scoring the corresponding comparison results according to the preset scoring rules to construct a judgment matrix; based on the maximum eigenvalue of the judgment matrix and its corresponding eigenvector, obtaining the weight of each key factor, and the weight reflects the importance of the corresponding key factor in the qualification certification of the historical certification target; using the key factors of the historical qualification certification with corresponding weights, the corresponding policy and regulatory data and the historical industry benchmark data as training data, using the results of whether the historical qualification certification was passed as labels, iteratively training the random forest model, and obtaining a trained qualification certification analysis model; The construction process of the management consulting analysis model includes: obtaining internal data and external data of multiple historical time periods of different enterprises, as well as historical SWOT analysis results of the corresponding enterprises in the corresponding historical time periods, wherein the internal data includes enterprise operation data and enterprise technical data, and the external data includes industry data of the industry to which the target enterprise belongs and data of competing enterprises with competitive relationships; using the standardized data after data standardization of the internal data and external data of each historical time period as training data, using the corresponding historical SWOT analysis results as labels, iteratively training the support vector machine, and obtaining a trained management consulting analysis model for predicting the SWOT analysis results of enterprises.

[0014] In a second aspect, a data processing method in the field of qualification certification and management consulting is provided, which is applied to a data management network including a data management platform and multiple network partitions, each network partition is a sub-management network of an industry type to which a configured management enterprise belongs, and each network partition includes multiple data processing systems and databases with the same structure. After receiving a first data processing request including an industry type and an enterprise identifier sent by a target enterprise, the data management platform verifies whether the target enterprise is a legal enterprise by inputting the enterprise identifier into a configured enterprise credit information system. If not, an indication of refusing to process the first data processing request is generated; if yes, a second data processing request including the enterprise identifier is sent to the target network partition corresponding to the industry type; the enterprise identifier is a unified social credit code or an enterprise name; the target network partition schedules each data processing system using a response time scheduling rule to determine the data processing system that processes the second data processing request, and the response time scheduling rule is a rule for selecting the data processing system with the shortest response time. The method includes: Sending a request input instruction to a user of the target enterprise based on the second data processing request, and receiving a data collection request for qualification certification or management consulting corresponding to the request input instruction; the data collection request includes a data type and an identifier of data to be collected; the data type includes a qualification certification type or a management consulting type; Using the domain knowledge graphs corresponding to different data types stored in the database, determine the target domain knowledge graph corresponding to the data type; Determine an entity node related to the data identifier in the target domain knowledge graph, and determine the data source to be collected based on the entity node; Collecting target data corresponding to the entity node from the data source to be collected; Based on the constructed data analysis model, the target data is analyzed to obtain data analysis results; the data analysis results include qualification assessment information or management diagnosis information.

[0015] In a third aspect, an electronic device is provided, the electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement any method step described in the first aspect when executing the program stored in the memory.

[0016] In a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any method step described in the first aspect is implemented.

[0017] In the data processing system and method in the field of qualification certification and management consulting provided by the embodiment of the present application, the system is applied to a data management network including a data management platform and multiple network partitions, each network partition is a sub-management network of an industry type to which the configured management enterprise belongs, and each network partition includes multiple data processing systems and databases with the same structure. After receiving the first data processing request including the industry type and enterprise identification sent by the target enterprise, the data management platform verifies whether the target enterprise is a legal enterprise by inputting the enterprise identification into the configured enterprise credit information system. If not, an indication of refusing to process the first data processing request is generated; if so, a second data processing request including the enterprise identification is sent to the target network partition corresponding to the industry type; the enterprise identification is a unified social credit code or an enterprise name; the target network partition uses a response time scheduling rule to schedule each data processing system to determine the data processing system that processes the second data processing request, and the response time scheduling rule is to select The rules of the data processing system with the shortest response time include: an interaction module, a data collection module and a data analysis module; wherein: the interaction module is used to send a request input instruction to a user of a target enterprise based on a second data processing request, and receive a data collection request for qualification certification or management consulting corresponding to the request input instruction; the data collection request includes a data type and a data identifier to be collected; the data type includes a qualification certification type or a management consulting type; the data collection module is used to use the domain knowledge graph corresponding to different data types stored in the database to determine the target domain knowledge graph corresponding to the data type, and determine the entity node related to the data identifier in the target domain knowledge graph, determine the data source to be collected based on the entity node, and collect the target data corresponding to the entity node from the data source to be collected; the data analysis module is used to analyze the target data based on the constructed data analysis model to obtain the data analysis result; the data analysis result includes qualification evaluation information or management diagnosis information. The system realizes the automation of data collection, and at the same time uses the domain knowledge graph corresponding to different data types stored in the database to ensure that the data collection and analysis process is based on an authoritative and professional knowledge system, improves the accuracy and reliability of the data, and reduces the risk of misjudgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of a data management network provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a data processing system provided in an embodiment of the present application; Figure 3 A flowchart of a data processing method in the field of qualification certification and management consulting provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] The data processing system in the field of qualification certification and management consulting provided in the embodiment of the present application can be applied to Figure 1 In the data management network shown, the data management network may include a data management platform and multiple network partitions, such as network partition 1, ..., network partition n, each network partition is a sub-management network of a configured management enterprise belonging to a certain industry type, that is, each network partition corresponds to a certain industry type (such as the construction industry, entertainment industry, film and television industry, etc.) to which the enterprise belongs, and each network partition may include multiple data processing systems and a database with the same structure for qualification certification and management consulting fields. Figure 1 Take the structure shown in network partition 1 as an example. Among them, multiple data processing systems with the same structure means that each data processing system is composed of the same modules, that is, they are the same in structure, but the data processed can be the same or different, that is, any two data processing systems can be data processing systems with the same structure and the same data processing, such as each data processing system can process both qualification certification and management consulting; or, any two data processing systems can be data processing systems with the same structure but different data processing, such as one part of the data processing system can process qualification certification, and the other part of the data processing system can process management consulting. The data management platform is connected to each network partition in communication, and the data processing systems in each network partition may or may not be connected in communication, according to actual needs. Each data processing system in each network partition is connected in communication with the database.

[0022] The data management platform receives a first data processing request sent by a target enterprise, which may include an industry type and an enterprise identifier, which may be a unified social credit code or enterprise name of the target enterprise. Afterwards, the data management platform verifies whether the target enterprise is a legal enterprise by inputting the enterprise identifier into a configured enterprise credit information system. If not, an instruction to refuse to process the first data processing request is generated; if so, a second data processing request including the enterprise identifier is sent to a target network partition corresponding to the industry type; wherein the configured enterprise credit information system may be China's National Enterprise Credit Information Publicity System, or an enterprise background investigation service system of a professional business information service provider (such as Tianyancha, Qichacha, etc.).

[0023] The target network partition uses the response time scheduling rule to schedule different data processing systems and determine the data processing system that processes the second data processing request. The response time scheduling rule is a rule for selecting the data processing system with the shortest response time from a limited number of data processing systems. Among them, using the response time scheduling rule to schedule the data processing system can improve the response speed and user experience of the overall system. Specifically, the response time data of each data processing system processing request is collected in real time. This can be achieved through embedded code, log analysis or dedicated monitoring software. In order to smooth short-term fluctuations and reflect long-term trends, the sliding window method can be used to calculate the average response time of each data processing system. For example, the average of all response times of each data processing system in the past 5 minutes is taken as the current average response time of the corresponding data processing system. Whenever a new request arrives, the average response time of all data processing systems is queried and sorted from low to high by response time. Then, the data processing system with the shortest response time is selected to process the request.

[0024] The industry-based network partitioning of this application can divide enterprises into different network partitions according to industry types, reducing the complexity of cross-industry data processing. Each network partition focuses on data processing tasks in a specific industry, improving the degree of specialization and processing efficiency.

[0025] In some embodiments, any two data processing systems can be data processing systems with the same structure and the same data processing. Since the communication quality will decrease with the increase of distance, the multiple network partitions included in the data management network can be divided according to geographical location, and each network partition is assigned a main data processing system and several backup data processing systems. When calling, the main data processing system of the partition is prioritized; if the main data processing system is unavailable (failure or insufficient resources), it is switched to the backup data processing system for processing; if the main data processing system is unavailable due to insufficient resources, the backup data processing system needs to send the processing results to the main data processing system after the processing is completed for feedback. If the main data processing system is unavailable due to a failure, the data processing system is still faulty after the backup data processing system is completed. At this time, the backup data processing system directly feedbacks the processing results.

[0026] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application may be combined with each other if there is no conflict.

[0027] like Figure 2 As shown, the data processing system may include an interaction module, a data acquisition module and a data analysis module, wherein: The interactive module is used to submit a data collection request for qualification certification or management consulting to the data collection module through the user interface of the target enterprise; the data collection request may include a data type and a data identifier to be collected; the data type may include a qualification certification type or a management consulting type, and the data identifier is a data name related to the data type, such as the data identifier is a name of a device facility related to the management consulting type. The interactive module provides a clear user interface to facilitate enterprise users to submit data collection requests. Through the interactive module, users can select data types (qualification certification type or management consulting type) according to actual needs, which enhances the flexibility and personalized service capabilities of the system.

[0028] A data collection module is used to determine the target domain knowledge graph corresponding to the data type by using the configured domain knowledge graphs corresponding to different data types, and determine the entity node related to the data identifier in the target domain knowledge graph, determine the data source to be collected based on the entity node, and collect the target data corresponding to the entity node from the data source to be collected; The data analysis module is used to analyze the target data based on the constructed data analysis model to obtain data analysis results; the data analysis results may include qualification assessment information or management diagnosis information, and the data analysis results are stored in the database.

[0029] Among them, the qualification assessment information may include relevant information that characterizes whether the target enterprise meets the requirements of the corresponding qualification level. For example, in the qualification certification of construction enterprises, it is necessary to analyze whether the enterprise meets the requirements of the corresponding qualification level in terms of personnel configuration (such as the number and qualifications of professional and technical personnel), equipment and facilities (such as the number and performance of construction machinery and equipment owned), and engineering performance (such as the scale and quality of completed engineering projects). Using the pre-configured qualification certification analysis model for construction enterprises, relevant target data can be calculated and compared. For example, the proportion of the number of professional and technical personnel owned by the enterprise to the total number of employees is calculated, and compared with the minimum proportion requirement in the qualification standard. Thus, it can be obtained whether the construction enterprise meets the personnel configuration of professional and technical personnel required by the corresponding qualification level.

[0030] Management diagnostic information can include relevant information that characterizes the internal management status and external market environment status of the target enterprise. For example, management theories and tools can be used to analyze the management data of the enterprise. For example, SWOT analysis (strengths, weaknesses, opportunities, and threats analysis) can be used to evaluate the internal management and external environment status of the enterprise, so as to analyze the financial health status (such as current ratio, debt-to-asset ratio and other indicators), human resource management efficiency (such as employee turnover rate, training coverage and other indicators), market competitiveness (such as market share, customer satisfaction and other indicators) of the enterprise to understand the internal management status and external market environment status. For example, through analysis, it is found that the input-output ratio of the enterprise in marketing is lower than the industry average, which may be a manifestation of problems in the marketing strategy.

[0031] Furthermore, the system may also include a visualization module for displaying the data analysis results.

[0032] Specifically, the visualization module can intuitively display the analysis results in the form of charts, graphs, etc. For example, a bar chart can be used to display the trend of the company's financial income changes in different years, and a radar chart can be used to display the scores of the company in various dimensions of qualification assessment.

[0033] It can be seen that the system provided by this application realizes the intelligent collection, analysis and display of data in the field of qualification certification and management consulting. Specifically, an intelligent data collection mechanism, a customized data analysis model and an intuitive result display method have been developed. Among them, the accuracy and efficiency of data collection are improved through the domain knowledge map in the intelligent data collection mechanism; the customized data analysis model can provide targeted analysis services according to different data types (such as qualification certification type or management consulting type); the intuitive result display method can help users quickly obtain valuable information, thereby realizing accurate enterprise qualification certification or management.

[0034] The functions of the above modules are described in detail below: (1) For the data acquisition module: In the specific implementation, first, configure the domain knowledge graphs corresponding to different data types, including: Since different data types include qualification certification type or management consulting type, the qualification certification type data or management consulting type data corresponding to the target enterprise can be obtained from the target enterprise's internal documents (such as employee files, equipment lists, project reports, corporate financial statements, human resources information, project management documents, etc.), industry standard documents (such as various qualification certification standards, management specifications, etc.), public databases (such as industry statistics published by servers, corporate credit information, industrial and commercial registration information, industry statistics, etc.) and third-party data (such as data from credit rating agencies) to configure the first domain knowledge graph corresponding to the qualification certification type and the second domain knowledge graph corresponding to the management consulting type.

[0035] The configuration of the first domain knowledge graph corresponding to the qualification certification type includes: selecting the first enterprise information (such as the industry attributes of enterprises such as construction enterprises and manufacturing enterprises), personnel information (such as professional and technical personnel, management personnel, etc.), equipment and facility information (such as mechanical equipment, testing instruments, etc.), qualification standard information (such as ISO9001, construction enterprise qualification level standards, etc.), project information (completed projects, ongoing projects, etc.), and the "employment" relationship between enterprise information and personnel information; the "ownership" relationship between enterprise information and equipment and facility information; the "conformance / non-conformance" relationship between enterprise information and qualification standard information, and the "construction" relationship between project information and enterprise information. Afterwards, the first domain knowledge graph is constructed with the target enterprise's industry attribute entities, personnel entities, equipment and facility entities, qualification standard entities, and project entities as nodes, and the relationships between different types of entities as edges.

[0036] The configuration of the second domain knowledge graph corresponding to the management consulting type includes: filtering out the second enterprise information (i.e., the enterprise receiving management consulting services), department information (such as human resources department, finance department, and other department categories), management process (such as recruitment process, performance appraisal process, etc.), management tools (such as ERP system, CRM system, etc.), management problems (such as high employee turnover, capital turnover difficulties, etc.) from the acquired management consulting type data, as well as the "execution" relationship between department information and management process; the "solution" relationship between management problems and management tools; and the "cooperation" relationship between enterprises and management consulting companies. Afterwards, the second domain knowledge graph is constructed with the consulting service entities, department entities, management process entities, management tool entities, and management problem entities of the target enterprise as nodes, and the relationships between different types of entities as edges.

[0037] In addition, it is necessary to record the data source of the data corresponding to the entity nodes in the domain knowledge graph. For example, the equipment and facility entities in the first domain knowledge graph correspond to equipment types such as mechanical equipment and testing instruments, and their data source is the equipment list.

[0038] Furthermore, the target domain knowledge graph corresponding to the data type in the data collection request is determined by utilizing the domain knowledge graphs corresponding to the configured different data types.

[0039] Secondly, determine the entity nodes related to data identification in the target domain knowledge graph, and determine the data source to be collected based on the entity nodes.

[0040] If the data identifier is an identifier of an equipment or facility type, then the entity node directly related to the data identifier can be determined in the corresponding domain knowledge graph as the initial entity directly related to the data identifier, such as an equipment or facility type entity, and then the data source containing relevant information of the initial entity (such as an equipment or facility type entity) is further searched, such as the company's equipment procurement records, equipment maintenance records, etc., that is, at this time, an entity node directly related to the data identifier and at least one data source corresponding to the corresponding entity node are determined.

[0041] In some embodiments, if there are multiple entity nodes related to the data identifier determined in the target domain knowledge graph, then the entity nodes related to the data identifier are determined in the target domain knowledge graph, including: determining the initial entity node corresponding to the data identifier in the target domain knowledge graph; counting the first entity node that has a relationship edge with the initial entity node, the second entity node that has two relationship edges, and the third entity node that has three relationship edges; and determining the initial entity node, the first entity node, the second entity node, and the third entity node as entity nodes related to the data identifier.

[0042] In a specific implementation, after determining the initial entity directly related to the data identifier, the relationship edge in the corresponding domain knowledge graph is used to determine the entity node related to the initial entity to determine the entity node indirectly related to the data identifier, that is, at this time, an entity node directly related to the data identifier and at least one data source corresponding to the corresponding entity node, as well as an entity node indirectly related to the data identifier and at least one data source corresponding to the corresponding entity node are determined. For example, in the second domain knowledge graph, through a multi-step relationship such as "enterprise-cooperation-management consulting company-use-management tool", potential data sources related to management tools can be found, such as user manuals and update logs provided by management tool suppliers. For another example, if the data identifier is a professional and technical personnel identifier, the node directly related to the professional and technical personnel identifier in the first domain knowledge graph is a professional and technical personnel entity in the personnel entity, thereby determining the data source related to the professional and technical personnel entity, and also based on the relationship between professional and technical personnel and qualification certification standards in the first domain knowledge graph, determine the data source including professional qualification examination scores, qualification certificate issuance records, etc.

[0043] Finally, corresponding collection strategies can be formulated according to different data sources to collect target data corresponding to entity nodes from the data sources to be collected. Among them, the collection strategy can include that for internal enterprise data, it can be connected with the company's existing ERP (enterprise resource planning), CRM (customer relationship management) and other systems by deploying data collection interfaces; for external public data, web crawler technology is used to capture it regularly according to certain rules; for third-party data, API (application programming interface) connection is established to clarify data request frequency, format requirements, etc.

[0044] (2) For the data analysis module: The data analysis module may include a qualification certification analysis model and a management consulting analysis model.

[0045] The construction of the qualification certification analysis model includes: determining the historical certification target of the enterprise qualification certification and the key factors affecting the historical certification target; obtaining the comparison results of experts comparing different key factors in pairs, and scoring the corresponding comparison results according to the preset scoring rules (such as 1-9 scaling method) to construct a judgment matrix; based on the maximum eigenvalue of the judgment matrix and its corresponding eigenvector, the weight of each key factor is obtained, and the weight reflects the importance of the corresponding key factor in the qualification certification of the historical certification target; then, the key factors of the historical qualification certification with corresponding weights, the corresponding policy and regulatory data and the historical industry benchmark data are used as training data, and the results of whether the historical qualification certification is passed or not are used as labels, and the random forest model is iteratively trained to obtain a trained qualification certification analysis model.

[0046] The construction of the management consulting analysis model includes: obtaining the internal and external data of different enterprises in multiple historical time periods, as well as the historical SWOT analysis results of the corresponding enterprises in the corresponding historical time periods. External threats can be evaluated by obtaining data from channels such as competitors' annual reports and news reports. After that, the standardized data after data standardization of the internal and external data of each historical time period is used as training data, and the corresponding historical SWOT analysis results are used as labels to iteratively train the support vector machine to obtain a trained management consulting analysis model for predicting the SWOT analysis results of the enterprise. Among them, the internal data can include enterprise operation data and enterprise technology data. The enterprise operation data can include financial statements (such as revenue, cost, profit, etc.), human resources data (number of employees, skill level distribution, etc.), and production data (output, quality pass rate, etc.). For example, when analyzing the strengths and weaknesses of a manufacturing company, a high profit margin in the financial statement may indicate that the company has a cost control advantage, while a low output may be a disadvantage. Enterprise technology data can be technical documents from the internal R&D department of the enterprise, patent application records, etc. These data help identify the unique capabilities or shortcomings of the enterprise in technology. External data can include industry data of the target company's industry and data of competing companies with whom the target company is competing; among them, industry data can include industry reports and market trend analysis data. For example, the growth rate data of the global new energy vehicle industry can be used as a data source for identifying opportunities. Competitive company data can include competitors' financial status, market share, new product release information, etc.

[0047] In some embodiments, when the data analysis module analyzes the target data based on the constructed data analysis model, it can configure the data priority corresponding to the entity node based on the number of relationship edges between the entity nodes related to the data identifier; based on the constructed data analysis model, the target data corresponding to each entity node is analyzed according to the data priority to obtain the data analysis results.

[0048] In some embodiments, the interaction module is further used to receive feedback information from different users through the user interface of the target enterprise, and the feedback information may include questions and / or improvement suggestions raised by different users on the displayed data analysis results; The data analysis module is also used to classify the feedback information received from different users using a preset classification algorithm to obtain at least one feedback information cluster; the feedback information in each feedback information cluster is input into the trained key information extraction model in sequence to obtain the key information corresponding to each feedback information cluster; the key information extraction model can be based on a feedback fragment training data set containing feedback fragments marked as key information and non-key information, and is trained by a classification model to be trained (such as a support vector machine, random forest, or other model with a classification algorithm); based on the key information extracted from each feedback information cluster and the correlation between different key information, an information knowledge graph representing the key information is constructed; wherein the information knowledge graph uses the key concept entities corresponding to the key information as nodes, and the correlation between the key information as edges between different key concept entities.

[0049] In some embodiments, the data analysis module specifically uses a preset classification algorithm to classify the feedback information received from different users to obtain at least one feedback information cluster, and the execution process includes: For the field of qualification certification or management consulting, a first feature extraction module of historical feedback information and a second feature extraction module of historical feedback information are determined. The first feature extraction module and the second feature extraction module have the same structure and different parameters. The first feature extraction module and the second feature extraction module are obtained by literature training based on historical feedback information. Among them, the network structure of the first feature extraction module and the second feature extraction module are exactly the same, and both can include: an input embedding layer and a 12-layer Transformer (a deep learning architecture) encoder. The input embedding layer is used to convert words and tags in the input text into vector representations. The 12-layer Transformer encoder is composed of 12 Transformer encoders stacked, and each Transformer encoder is composed of multiple self-attention mechanism layers and feedforward neural network layers. However, since the first feature extraction module and the second feature extraction module have different parameter update rules during the training process, the parameters of the trained modules finally obtained through training optimization are different.

[0050] Among them, there are multiple options for the parameter update rules of the first feature extraction module and the second feature extraction module according to actual needs, and the parameter update rules of the two modules can complement each other. For example, the parameter update of the first feature extraction module can focus more on learning and optimizing the results, and the parameter update of the second feature extraction module can focus more on the inheritance of process experience.

[0051] As an example, the parameters of the first feature extraction module are updated based on the first rule; the second feature extraction module is updated based on the second rule. The first rule is updated based on a predefined loss function, and the second rule is updated based on the historical parameters of the first feature extraction module. The definitions of the first rule and the second rule are described in detail later and will not be repeated here. By training the two modules, it is possible to enhance the feature extraction capability of the modules and enhance the module's ability to recognize comprehensive and applied feedback information.

[0052] Afterwards, the feedback information currently received from different users can be input into the first feature extraction module to obtain a first dense feature vector set for each feedback information, and the first dense feature vector set can be input into a predetermined similarity scoring module to obtain a first similarity score; the feedback information currently received from different users can be input into the second feature extraction module to obtain a second dense feature vector set for each feedback information, and the second dense feature vector set can be input into a predetermined similarity scoring module to obtain a second similarity score.

[0053] Specifically, for the field of qualification certification or the field of management consulting, a first feature extraction module of historical feedback information and a second feature extraction module of historical feedback information are determined, the first feature extraction module and the second feature extraction module have the same structure and different parameters, and the first feature extraction module and the second feature extraction module are obtained by training based on historical feedback information; the feedback information in the current cycle is input into the first feature extraction module to obtain a first dense feature vector set of each feedback information, and the first dense feature vector set is input into a predetermined similarity scoring module to obtain a first similarity score; the feedback information in the current cycle is input into the second feature extraction module to obtain a second dense feature vector set of each feedback information, and the second dense feature vector set is input into a predetermined similarity scoring module to obtain a second similarity score; the first similarity score and the second similarity score of each feedback information are fused to obtain a fused similarity score of each feedback information; based on the fused similarity score, the first feedback information set belonging to the field of qualification certification and the second feedback information set belonging to the field of management consulting are determined in the feedback information in the current cycle to be classified; wherein the fused similarity score is used to evaluate the similarity between the feedback information in the current cycle to be classified and the feedback information in the field of qualification certification or the feedback information in the field of management consulting.

[0054] Specifically, the first similarity score and the second similarity score The calculation formula can be expressed as: ; ; In the formula, the first dense feature vector set is , the second dense feature vector set is , n is the number of feedback information in the current cycle, i∈n, g is the average vector of feedback information in the field of qualification certification or feedback information in the field of management consulting, where, is the shared covariance matrix of the feedback information in the field of qualification certification and the feedback information in the field of management consulting, for The pseudo-inverse matrix of ; the fusion similarity score is determined based on the following formula: ;in, is the weight coefficient. The size of can be an empirical value, or it can be optimized according to the quality of the classification results. The judgment of the quality of the classification results here can be verified using a validation set with known labels.

[0055] The size of the fusion similarity score is used to indicate the probability that the document belongs to the in-domain document.

[0056] Afterwards, based on the fusion similarity score, target feedback information belonging to the field of qualification certification or the field of management consulting is determined among the feedback information of different users currently received to be classified.

[0057] A first sample is generated for each first feedback information set, and a corresponding second sample is generated for each first feedback information using a graph encoder; a third sample is generated for each second feedback information set, and a corresponding fourth sample is generated for each second feedback information using a graph encoder; the second sample corresponding to the same first feedback information is a positive sample of the first sample, the second sample, the third sample and the fourth sample corresponding to different first feedback information are negative samples of the first sample, the fourth sample corresponding to the same second feedback information is a positive sample of the third sample, and the fourth sample, the first sample and the second sample corresponding to different second feedback information are negative samples of the first sample; the parameters in the first feature extraction module are optimized by narrowing the difference between the first sample or the third sample and the positive sample, and by widening the difference between the first sample or the third sample and the negative sample, and the parameters of the second feature extraction module are optimized based on the historical parameters of the first feature extraction module, so as to obtain the first feature extraction module of the current period and the second feature extraction module of the current period; the feedback information in the current period is classified based on the first feature extraction module and the classifier of the current period to obtain at least one feedback information cluster.

[0058] Among them, the graph encoder is a tool used to encode the label hierarchy and output label features. This application adopts Graphormer. Graphormer introduces the Transformer module in the encoding of the graph and shows good performance on multiple graph-related tasks. The uniqueness of Graphormer is that in addition to introducing the Transformer module, three encoding methods are also adopted, which greatly enhances the representation ability of the graph. The first is centrality encoding, which is used to capture the importance of nodes in the graph. By assigning a learnable vector to each node, the vector is added to the node features of the input layer. The second is spatial encoding, which assigns learnable embeddings to node pairs according to the spatial relationship between nodes. The third is edge encoding, which focuses on processing additional spatial information in edge features. This encoding method calculates the dot product mean of the edge features and the learnable embedding on the shortest path, and uses it in the Attention module. Overall, Graphormer models graphs from multiple perspectives and can be easily applied to hierarchical text classification tasks.

[0059] Furthermore, the visualization display module is also used to display the frequency distribution of key information extracted from different feedback information clusters using preset display graphics, and to display the information knowledge graph using preset network graphics.

[0060] Furthermore, the system may also include a report generation module; the report generation module is used to automatically generate a detailed report based on the data analysis results. The report content may include a data overview, analysis process, problems found, improvement suggestions, etc. For example, in terms of qualification certification, the report content may list in which aspects the enterprise meets the qualification requirements, in which aspects there are deficiencies, and provide specific improvement measures; in terms of management consulting, the report content may describe in detail the management status of the enterprise, the existing problems, and the solutions proposed for these problems.

[0061] The system of this application provides enterprises with comprehensive and efficient data management and analysis services through the following core advantages: Efficiency: Network partitioning and dynamic scheduling mechanism based on industry types significantly improve data processing efficiency. Security: Strict legality verification and application of domain knowledge graphs ensure data security and accuracy. Flexibility: Modular design and user-defined functions meet diverse needs. Applicability: Supports multi-industry applications and has strong scalability and customization capabilities.

[0062] Based on the above data processing system, this application also provides a data processing method in the field of qualification certification and management consulting, such as Figure 3 As shown, the method may include: Step 310: Send a request input instruction to a user of the target enterprise based on the second data processing request, and receive a data collection request for qualification certification or management consulting corresponding to the request input instruction.

[0063] The data collection request may include a data type and an identifier of the data to be collected; the data type may include a qualification certification type and a management consulting type; wherein the acquisition process of the second data processing request is: After receiving the first data processing request including the industry type and enterprise identification sent by the target enterprise, the data management platform in the data management network sends a second data processing request including the enterprise identification to the target network partition corresponding to the industry type if it is verified that the target enterprise is a legal enterprise; the target network partition is scheduled based on the working status of each data processing system, and determines the data processing system that processes the second data processing request so that the data processing system receives the second data processing request.

[0064] Step 320: Utilize the domain knowledge graphs corresponding to different data types stored in the database to determine the target domain knowledge graph corresponding to the data type, determine the entity nodes related to the data identifier in the target domain knowledge graph, determine the data source to be collected based on the entity nodes, and collect the target data corresponding to the entity nodes from the data source to be collected.

[0065] Step 330: Analyze the target data based on the constructed data analysis model to obtain data analysis results.

[0066] Data analysis results include qualification assessment information or management diagnosis information.

[0067] Optionally, the domain knowledge graphs corresponding to the different data types include a first domain knowledge graph corresponding to the qualification certification field, and a second domain knowledge graph corresponding to the management consulting field; wherein: the first domain knowledge graph is a knowledge graph constructed with the target enterprise's industry attribute entities, personnel entities, equipment and facility entities, qualification standard entities and engineering project entities as nodes, and the relationships between different types of entities as edges; the second domain knowledge graph is a knowledge graph constructed with the target enterprise's consulting service entities, department entities, management process entities, management tool entities and management problem entities as nodes, and the relationships between different types of entities as edges.

[0068] Optionally, when there are multiple entity nodes related to the data identifier determined in the target domain knowledge graph, the initial entity node corresponding to the data identifier in the target domain knowledge graph can be determined; the first entity node with one relationship edge between the initial entity node, the second entity node with two relationship edges and the third entity node with three relationship edges are counted; the initial entity node, the first entity node, the second entity node and the third entity node are determined as entity nodes related to the data identifier.

[0069] Optionally, the data priority corresponding to the entity node can be configured based on the number of relationship edges between the entity nodes related to the data identifier; based on the constructed data analysis model, the target data corresponding to each entity node is analyzed according to the data priority to obtain the data analysis result.

[0070] Optionally, feedback information from different users can be received through the user interface of the target enterprise, the feedback information including questions and / or improvement suggestions raised by the different users regarding the displayed data analysis results; a preset classification algorithm is used to classify the feedback information received from different users to obtain at least one feedback information cluster; the feedback information in each feedback information cluster is input into a trained key information extraction model in sequence to obtain key information corresponding to each feedback information cluster, the key information extraction model being obtained by training a classification model to be trained based on a training data set of feedback fragments containing feedback fragments marked as key information and non-key information; an information knowledge graph representing the key information is constructed based on the key information extracted from each feedback information cluster and the association relationship between different key information; the information knowledge graph uses key concept entities corresponding to the key information as nodes, and the association relationship between the key information as edges between different key concept entities.

[0071] Optionally, the data analysis results are displayed, and the frequency distribution of key information extracted from different feedback information clusters is displayed using a preset display graphic, and the information knowledge graph is displayed using a preset network graphic.

[0072] Optionally, the constructed data analysis model includes a qualification certification analysis model and a management consulting analysis model; The construction process of the qualification certification analysis model includes: determining the historical certification target of the enterprise qualification certification and the key factors affecting the historical certification target; obtaining the comparison results of the experts comparing different key factors in pairs, and scoring the corresponding comparison results according to the preset scoring rules to construct a judgment matrix; based on the maximum eigenvalue of the judgment matrix and its corresponding eigenvector, obtaining the weight of each key factor, and the weight reflects the importance of the corresponding key factor in the qualification certification of the historical certification target; using the key factors of the historical qualification certification with corresponding weights, the corresponding policy and regulatory data and the historical industry benchmark data as training data, using the results of whether the historical qualification certification was passed as labels, iteratively training the random forest model, and obtaining a trained qualification certification analysis model; The construction process of the management consulting analysis model includes: obtaining internal data and external data of multiple historical time periods of different enterprises, as well as historical SWOT analysis results of the corresponding enterprises in the corresponding historical time periods, wherein the internal data includes enterprise operation data and enterprise technical data, and the external data includes industry data of the industry to which the target enterprise belongs and data of competing enterprises with competitive relationships; using the standardized data after data standardization of the internal data and external data of each historical time period as training data, using the corresponding historical SWOT analysis results as labels, iteratively training the support vector machine, and obtaining a trained management consulting analysis model for predicting the SWOT analysis results of enterprises.

[0073] Each step of the data processing method in the field of qualification certification and management consulting provided by the above-mentioned embodiments of the present application can be implemented through the functions of the above-mentioned modules. Therefore, the specific implementation process and beneficial effects of each step in the data processing method in the field of qualification certification and management consulting provided by the embodiments of the present application will not be repeated here.

[0074] The present application also provides an electronic device, such as Figure 4 As shown, it includes a processor 410 , a communication interface 420 , a memory 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other via the communication bus 440 .

[0075] Memory 430, for storing computer programs; The processor 410 is used to execute the program stored in the memory 430 to implement the following steps: Sending a request input instruction to a user of the target enterprise based on the second data processing request, and receiving a data collection request for qualification certification or management consulting corresponding to the request input instruction; the data collection request includes a data type and an identifier of data to be collected; the data type includes a qualification certification type and a management consulting type; Using the domain knowledge graphs corresponding to different data types stored in the database, determine the target domain knowledge graph corresponding to the data type; Determine an entity node related to the data identifier in the target domain knowledge graph, and determine the data source to be collected based on the entity node; Collecting target data corresponding to the entity node from the data source to be collected; Based on the constructed data analysis model, the target data is analyzed to obtain data analysis results; the data analysis results include qualification assessment information or management diagnosis information.

[0076] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0077] The communication interface is used for communication between the above electronic device and other devices.

[0078] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0079] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0080] The implementation methods and beneficial effects of the components of the electronic device in the above embodiments to solve the problems can be seen in Figure 2The various steps in the illustrated embodiment are implemented, therefore, the specific working process and beneficial effects of the electronic device provided by the embodiment of the present application are not repeated here.

[0081] In another embodiment provided in the present application, a computer-readable storage medium is also provided, in which instructions are stored. When the computer-readable storage medium is executed on a computer, the computer executes the data processing method in the field of qualification certification and management consulting described in any of the above embodiments.

[0082] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute the data processing method in the field of qualification certification and management consulting described in any one of the above embodiments.

[0083] Those skilled in the art will appreciate that the embodiments in the present application can be provided as methods, systems, or computer program products. Therefore, the embodiments in the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments in the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0084] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0087] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.

[0088] Obviously, those skilled in the art can make various changes and modifications to the embodiments in the present application without departing from the spirit and scope of the embodiments in the present application. Thus, if these modifications and variations of the embodiments in the present application fall within the scope of the claims and their equivalents in the embodiments of the present application, the embodiments of the present application are also intended to include these modifications and variations.

Claims

1. A data processing system in the field of qualification certification and management consulting, characterized in that: Applied to a data management network including a data management platform and multiple network partitions, each network partition is a sub-management network of an industry type to which a configured management enterprise belongs, and each network partition includes multiple data processing systems and databases with the same structure. After receiving a first data processing request including an industry type and an enterprise identifier sent by a target enterprise, the data management platform verifies whether the target enterprise is a legal enterprise by inputting the enterprise identifier into a configured enterprise credit information system; if not, an instruction to refuse to process the first data processing request is generated; if yes, a second data processing request including the enterprise identifier is sent to the target network partition corresponding to the industry type; the enterprise identifier is a unified social credit code or an enterprise name; The target network partition schedules each data processing system using a response time scheduling rule to determine the data processing system that processes the second data processing request, wherein the response time scheduling rule is a rule for selecting the data processing system with the shortest response time, and the system includes: an interaction module, a data acquisition module, and a data analysis module; An interaction module is used to send a request input instruction to a user of a target enterprise based on the second data processing request, and receive a data collection request for qualification certification or management consulting corresponding to the request input instruction; the data collection request includes a data type and a data identifier to be collected; the data type includes a qualification certification type or a management consulting type; the data identifier is a data name related to the data type; A data collection module is used to determine the target domain knowledge graph corresponding to the data type by using the domain knowledge graphs corresponding to different data types stored in the database, and determine the entity node related to the data identifier in the target domain knowledge graph, determine the data source to be collected based on the entity node, and collect the target data corresponding to the entity node from the data source to be collected; The data analysis module is used to analyze the target data based on the constructed data analysis model to obtain data analysis results; the data analysis results include qualification assessment information or management diagnosis information.

2. The system according to claim 1, characterized in that The domain knowledge graphs corresponding to the different data types include a first domain knowledge graph corresponding to the qualification certification field, and a second domain knowledge graph corresponding to the management consulting field; wherein: The first domain knowledge graph is constructed with the target enterprise's industry attribute entities, personnel entities, equipment and facility entities, qualification standard entities, and engineering project entities as nodes, and the relationships between entities of different types as edges; The second domain knowledge graph is constructed with the consulting service entities, department entities, management process entities, management tool entities and management problem entities of the target enterprise as nodes and the relationships between different types of entities as edges.

3. The system according to claim 2, characterized in that When there are multiple entity nodes related to the data identifier determined in the target domain knowledge graph, the data collection module is used to: Determine the initial entity node corresponding to the data identifier in the target domain knowledge graph; Count the first entity nodes that have one relationship edge with the initial entity node, the second entity nodes that have two relationship edges with the initial entity node, and the third entity nodes that have three relationship edges with the initial entity node; The initial entity node, the first entity node, the second entity node and the third entity node are determined as entity nodes associated with the data identifier.

4. The system according to claim 3, characterized in that The data analysis module is specifically used for: Based on the number of relationship edges between entity nodes related to the data identifier, configure the data priority corresponding to the entity node; Based on the constructed data analysis model, the target data corresponding to each entity node is analyzed according to the data priority to obtain a data analysis result.

5. The system according to claim 1, wherein: The interaction module is further used to receive feedback information from different users through the user interface of the target enterprise, wherein the feedback information includes questions and / or improvement suggestions raised by the different users on the displayed data analysis results; The data analysis module is further used to classify the feedback information received from different users using a preset classification algorithm to obtain at least one feedback information cluster; Input the feedback information in each feedback information cluster into the trained key information extraction model in sequence to obtain the key information corresponding to each feedback information cluster, wherein the key information extraction model is obtained by training the classification model to be trained based on a training data set of feedback segments containing feedback segments marked as key information and non-key information; Based on the key information extracted from each feedback information cluster and the association relationship between different key information, an information knowledge graph representing the key information is constructed; the information knowledge graph uses the key concept entities corresponding to the key information as nodes, and the association relationship between the key information as edges between different key concept entities.

6. The system according to claim 5, characterized in that The data analysis module specifically adopts a preset classification algorithm to classify the feedback information received from different users to obtain at least one feedback information cluster, and the execution process includes: For the field of qualification certification or management consulting, determine a first feature extraction module of historical feedback information and a second feature extraction module of historical feedback information, wherein the first feature extraction module and the second feature extraction module have the same structure but different parameters, and the first feature extraction module and the second feature extraction module are obtained by training based on historical feedback information; Inputting the feedback information in the current cycle into the first feature extraction module to obtain a first dense feature vector set of each feedback information, and inputting the first dense feature vector set into a predetermined similarity scoring module to obtain a first similarity score; Inputting the feedback information in the current cycle into the second feature extraction module to obtain a second dense feature vector set of each feedback information, and inputting the second dense feature vector set into a predetermined similarity scoring module to obtain a second similarity score; Merging the first similarity score and the second similarity score of each feedback information to obtain a fused similarity score of each feedback information; Based on the fusion similarity score, determining a first feedback information set belonging to the field of qualification certification and a second feedback information set belonging to the field of management consulting in the feedback information to be classified in the current cycle; A first sample is generated for each set of the first feedback information, and a corresponding second sample is generated for each set of the first feedback information using a graph encoder; a third sample is generated for each set of the second feedback information, and a corresponding fourth sample is generated for each set of the second feedback information using a graph encoder; the second sample corresponding to the same first feedback information is a positive sample of the first sample, the second sample, the third sample and the fourth sample corresponding to different first feedback information are negative samples of the first sample, the fourth sample corresponding to the same second feedback information is a positive sample of the third sample, and the fourth sample, the first sample and the second sample corresponding to different second feedback information are negative samples of the first sample; the parameters in the first feature extraction module are optimized by narrowing the difference between the first sample or the third sample and the positive sample, and by widening the difference between the first sample or the third sample and the negative sample, and the parameters of the second feature extraction module are optimized based on the historical parameters of the first feature extraction module, so as to obtain the first feature extraction module of the current period and the second feature extraction module of the current period; the feedback information in the current period is classified based on the first feature extraction module and the classifier of the current period to obtain at least one feedback information cluster.

7. The system according to claim 6, characterized in that The fusion similarity score is used to evaluate the similarity between the feedback information to be classified in the current cycle and the feedback information in the field of qualification certification or the feedback information in the field of management consulting.

8. The system according to claim 5, characterized in that The system also includes a visualization display module for displaying the data analysis results, and using preset display graphics to display the frequency distribution of key information extracted from different feedback information clusters, and using preset network graphics to display the information knowledge graph.

9. The system according to claim 1, characterized in that The constructed data analysis model includes a qualification certification analysis model and a management consulting analysis model; The construction process of the qualification certification analysis model includes: determining the historical certification target of the enterprise qualification certification and the key factors affecting the historical certification target; obtaining the comparison results of the experts comparing different key factors in pairs, and scoring the corresponding comparison results according to the preset scoring rules to construct a judgment matrix; based on the maximum eigenvalue of the judgment matrix and its corresponding eigenvector, obtaining the weight of each key factor, and the weight reflects the importance of the corresponding key factor in the qualification certification of the historical certification target; using the key factors of the historical qualification certification with corresponding weights, the corresponding policy and regulatory data and the historical industry benchmark data as training data, using the results of whether the historical qualification certification was passed as labels, iteratively training the random forest model, and obtaining a trained qualification certification analysis model; The construction process of the management consulting analysis model includes: obtaining internal data and external data of multiple historical time periods of different enterprises, as well as historical SWOT analysis results of the corresponding enterprises in the corresponding historical time periods, wherein the internal data includes enterprise operation data and enterprise technical data, and the external data includes industry data of the industry to which the target enterprise belongs and data of competing enterprises with competitive relationships; using the standardized data after data standardization of the internal data and external data of each historical time period as training data, using the corresponding historical SWOT analysis results as labels, iteratively training the support vector machine, and obtaining a trained management consulting analysis model for predicting the SWOT analysis results of enterprises.

10. A data processing method in the field of qualification certification and management consulting, characterized in that: Applied to a data management network including a data management platform and multiple network partitions, each network partition is a sub-management network of an industry type to which a configured management enterprise belongs, and each network partition includes multiple data processing systems and databases with the same structure. After receiving a first data processing request including an industry type and an enterprise identifier sent by a target enterprise, the data management platform verifies whether the target enterprise is a legal enterprise by inputting the enterprise identifier into a configured enterprise credit information system. If not, an instruction to refuse to process the first data processing request is generated; if yes, a second data processing request including the enterprise identifier is sent to the target network partition corresponding to the industry type; the enterprise identifier is a unified social credit code or an enterprise name; The target network partition schedules each data processing system using a response time scheduling rule to determine the data processing system that processes the second data processing request, wherein the response time scheduling rule is a rule for selecting the data processing system with the shortest response time, and the method includes: Sending a request input instruction to a user of the target enterprise based on the second data processing request, and receiving a data collection request for qualification certification or management consulting corresponding to the request input instruction; the data collection request includes a data type and a data identifier to be collected; the data type includes a qualification certification type or a management consulting type; the data identifier is a data name related to the data type; Using the domain knowledge graphs corresponding to different data types stored in the database, determine the target domain knowledge graph corresponding to the data type; Determine an entity node related to the data identifier in the target domain knowledge graph, and determine the data source to be collected based on the entity node; Collecting target data corresponding to the entity node from the data source to be collected; Based on the constructed data analysis model, the target data is analyzed to obtain data analysis results; the data analysis results include qualification assessment information or management diagnosis information.

Images