An organization establishment evaluation method and system based on a large language model

Abstract

This invention provides a method and system for evaluating organizational structure and staffing based on a large language model. The method includes: collecting industry standard data and evaluation reference data related to organizational structure and staffing; constructing scoring rules for multiple preset dimensions based on the industry standard data and evaluation reference data; obtaining a level judgment threshold for each preset dimension; constructing an evaluation standard knowledge base based on the scoring rules and the level judgment thresholds; collecting data from multiple set data sources to obtain multi-source data of the target organization; using a large language model to perform semantic extraction on the multi-source data to obtain the organizational entities within the target organization and the relationships between them, as well as the operational feedback data of the target organization; and using the evaluation standard database to evaluate the operational feedback data to obtain the evaluation result of the organizational structure and staffing system of the target organization. The technical solution of this invention can improve the accuracy and efficiency of organizational structure and staffing evaluation.

Description

Technical Field

[0001] This invention relates to the field of organizational structure assessment technology, and in particular to an organizational structure assessment method and system based on a large language model. Background Technology

[0002] The rationality of organizational structure and staffing is a crucial factor in ensuring organizational efficiency and optimizing resource allocation. Current methods for evaluating the rationality of organizational structure and staffing involve expert review, questionnaires, focus groups, and statistical analysis, assessing the rationality of the structure and staffing based on the organization's actual operations. However, this evaluation method is easily influenced by the subjective opinions of experts and other personnel, failing to guarantee the accuracy of the evaluation results. Furthermore, the evaluation process is time-consuming and inefficient. Summary of the Invention

[0003] This invention provides a method and system for evaluating organizational structure based on a large language model, which is used to improve the accuracy of organizational structure evaluation and increase the efficiency of the evaluation process.

[0004] Specifically, this invention provides a method for institutional staffing assessment based on a large language model, comprising: Collect industry standard data and evaluation reference data related to institutional establishment, and construct scoring rules with multiple preset dimensions based on the industry standard data and the evaluation reference data; Obtain the level judgment threshold for each of the preset dimensions, and construct an evaluation standard knowledge base based on the scoring rules and the level judgment threshold; Data is collected from multiple set data sources to obtain multi-source data of the target organization, and a large language model is used to extract semantics from the multi-source data to obtain the organizational entities in the target organization and the relationships between them, as well as the operational feedback data of the target organization. The organizational structure of the target organization is constructed based on the organizational entities and the relationships between them, and the organizational structure is evaluated based on the departmental performance data using the scoring rules, so as to obtain the evaluation score of the organizational structure in each of the preset dimensions. Based on each evaluation score and its corresponding level judgment threshold, the merit level of the organizational structure in each preset dimension is obtained.

[0005] Furthermore, the step of constructing scoring rules with multiple preset dimensions based on the industry standard data and the evaluation reference data includes: Based on the industry standard data and the evaluation reference data, scoring rules are obtained for performance of duties, effectiveness of indicators, process efficiency, resource allocation, public satisfaction, and compliance risks.

[0006] Furthermore, after the step of obtaining the merit level of the organization's structure in each preset dimension based on each evaluation score and its corresponding level judgment threshold, the method further includes: A staffing assessment report for the target organization is generated based on each assessment score, and the staffing assessment results of the target organization are visualized based on the staffing assessment report.

[0007] Furthermore, the step of visually presenting the assessment results of the target organization based on the assessment report includes: Based on the assessment report, generate a radar chart of the organization's organizational score, a heat map of organizational issues, and / or a trend analysis chart of organizational resource utilization for the target organization.

[0008] Furthermore, prior to the step of collecting data from multiple designated data sources to obtain multi-source data of the target organization, the method further includes: Determine whether the target organization triggers the preset staffing evaluation criteria; If so, then the step of collecting data from multiple set data sources to obtain multi-source data of the target organization is performed.

[0009] Furthermore, after the step of obtaining the merit level of the organization's structure in each preset dimension based on each evaluation score and its corresponding level judgment threshold, the method further includes: Each of the assessment scores is weighted and fused to obtain a comprehensive assessment result for the target institution.

[0010] Furthermore, after the step of constructing the organizational structure of the target organization based on the organizational entities and their relationships, the method further includes: Construct a staffing knowledge graph of the target organization based on the aforementioned staffing system.

[0011] Furthermore, before the step of semantic extraction of the multi-source data using a large language model, the method further includes: preprocessing the multi-source data, wherein the preprocessing includes at least one of cleaning, deduplication, format standardization, and privacy desensitization.

[0012] Furthermore, after the step of constructing the evaluation standard knowledge base based on the scoring rules and the level judgment threshold, the method further includes: Determine whether any modifications to the evaluation rules have been received; If so, the evaluation criteria knowledge base will be updated according to the modifications.

[0013] On the other hand, the present invention also provides an organizational structure assessment system based on a large language model, which includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the organizational structure assessment method based on a large language model as described above.

[0014] The technical solution of this invention constructs scoring rules for multiple preset dimensions based on industry standard data and evaluation reference data related to organizational structure and staffing, thereby building an evaluation standard knowledge base based on multiple preset dimensions. After collecting multi-source data of the target organization from multiple set data sources, the institutional entities of the target organization are extracted from the multi-source data, and the organizational structure and staffing system of the target organization is constructed. Finally, the evaluation standard knowledge base is used to evaluate the organizational structure and staffing system to obtain the merit level of the organizational structure and staffing system in each preset dimension, thus completing the organizational structure and staffing assessment of the target organization. Since the technical solution of this invention collects multi-source data of the target organization from multiple set data sources, the accuracy and reliability of the obtained organizational structure and staffing system and the operational feedback data of the target organization can be guaranteed. Furthermore, the use of the evaluation standard knowledge base can evaluate the organizational structure and staffing system of the target organization from multiple preset dimensions, which can improve the comprehensiveness, accuracy, and reliability of the evaluation of the organizational structure and staffing system, thereby ensuring the accuracy of the organizational structure and staffing assessment of the target organization. Moreover, the technical solution of this invention does not require manual construction of evaluation standards and the organizational structure and staffing system of the target organization, thus improving the efficiency of the organizational structure and staffing assessment of the target organization.

[0015] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0016] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a flowchart illustrating an institutional staffing assessment method based on a large language model according to an embodiment of the present invention; Figure 2 This is a schematic diagram of an institutional staffing assessment method based on a large language model according to an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the assessment of organizational structure in an organizational structure assessment method according to an embodiment of the present invention; Figure 4 This is a radar chart of the target organization evaluation results in an organization establishment evaluation method according to an embodiment of the present invention; Figure 5This is a line graph of the assessment results of the target organization in an organizational structure assessment method according to an embodiment of the present invention; Figure 6 This is a knowledge graph of the organizational structure system of the target organization in an organizational structure assessment method according to an embodiment of the present invention; Figure 7 This is a schematic diagram of an institutional staffing assessment system based on a large language model according to an embodiment of the present invention. Detailed Implementation

[0017] The following reference Figures 1 to 7 This invention describes a method and system for organizational structure assessment based on a large language model, according to embodiments of the present invention. In this description, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" and "second" may explicitly or implicitly include at least one of that feature, that is, include one or more of that feature. In the description of the present invention, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. When a feature "includes or contains" one or more of the features it encompasses, unless otherwise specifically described, this indicates that other features are not excluded and may be further included.

[0018] In the description of this embodiment, the terms "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0019] Please see Figure 1 , Figure 1 The diagram shown is a flowchart of an institutional staffing assessment method based on a large language model, according to one embodiment of the present invention. The principle of this method is as follows: Figure 2 As shown, an evaluation standard knowledge base can be constructed based on policy documents, historical assessment data, and expert opinions. A large language model is then used to perform semantic recognition on the multi-source heterogeneous data of the target institution to obtain the institutional staffing system of the target institution. Then, the evaluation standard knowledge base is used to conduct intelligent evaluation of the institutional staffing system from multiple dimensions to improve the accuracy and reliability of the institutional staffing evaluation.

[0020] Specifically, the organizational structure assessment method based on a large language model in this embodiment includes the following steps: Step S101: Collect industry standard data and evaluation reference data related to the organization's establishment, and construct scoring rules for multiple preset dimensions based on the industry standard data and evaluation reference data; Step S102: Obtain the level judgment threshold for each preset dimension, and construct an evaluation standard knowledge base based on the scoring rules and level judgment thresholds for each preset dimension. Step S103: Collect data from multiple set data sources to obtain multi-source data of the target organization, and use a large language model to extract semantics from the multi-source data to obtain the organizational entities of the target organization and the relationships between them, as well as the operational feedback data of the target organization. Step S104: Construct the organizational structure system of the target organization based on the organizational entities of the target organization and the relationships between them, and evaluate the organizational structure system based on the operational feedback data of the target organization using each scoring rule to obtain the evaluation score of the organizational structure system in each preset dimension. Step S105: Based on the level judgment threshold of each preset dimension and the evaluation score of the organizational structure system of the target organization in each preset dimension, the quality level of the organizational structure system of the target organization is obtained.

[0021] In step S101 above, the industry standard data related to organizational structure and staffing include laws, regulations, policy documents and industry standard documents related to organizational structure and staffing, and the assessment reference data related to organizational structure and staffing include the guiding assessment system of organizational structure and staffing departments and the assessment experience of experts.

[0022] In this embodiment, the acquired scoring rules for multiple preset dimensions include scoring rules for responsibility performance, indicator effectiveness, process efficiency, resource allocation, public satisfaction, and compliance risk. The scoring rule for responsibility performance uses semantic similarity calculation and logical reasoning to align and match the actual work content of the target organization with the legally mandated responsibilities stipulated in industry standard data. This quantitatively assesses the completeness, accuracy, and effectiveness of the target organization's responsibility performance, resulting in an evaluation score for the target organization's organizational structure in the responsibility performance dimension. This score is used to identify issues such as responsibilities being missing, overstepping, or misaligned within the organizational structure. The scoring rule for organizational effectiveness analyzes the target organization's personnel quantity and structure (age, education, major, etc.) and assesses the matching degree between this data and the workload and complexity of the business to evaluate the suitability of personnel for positions and workload, resulting in an evaluation score for the target organization's organizational structure in the organizational effectiveness dimension. The scoring rule for process efficiency analyzes the target organization's business processes and actual operational data to identify redundant links, bottlenecks, and cross-departmental collaboration obstacles, resulting in an evaluation score for the target organization's organizational structure in the process efficiency dimension. The resource allocation scoring rule assesses the rationality and efficiency of the allocation of human resources, budget, and equipment among departments and projects within a target organization, thereby obtaining an evaluation score for the organization's organizational structure in the resource allocation dimension. The public satisfaction scoring rule utilizes sentiment analysis and topic modeling to mine data from the target organization's complaints, public opinion, and service evaluations, quantifying public satisfaction to obtain an evaluation score for the organization's organizational structure in the public satisfaction dimension. The compliance risk scoring rule examines whether the target organization's operations comply with legal, regulatory, and policy requirements, identifying compliance loopholes and management risks to obtain an evaluation score for the organization's organizational structure in the compliance risk dimension.

[0023] In step S102 above, multiple quality levels can be pre-set for each preset dimension, and a corresponding quality judgment threshold can be set for each quality level. After obtaining the standard score of the target organization in each preset dimension, the quality level of the target organization in each preset dimension can be determined according to the threshold range in which each standard score falls.

[0024] In step S103 above, the target organization's multiple data sources include organizational regulations, organizational structure assessment self-assessment reports, work summaries and plans, internal rules and regulations, business process documents, meeting minutes, financial reports, personnel files, public service data, petition and hotline records, news reports, social media public opinion, and relevant policies and regulations.

[0025] This embodiment employs a large language model to extract semantic information from the collected data. This includes: using a large language model to perform semantic recognition on institutional regulations, self-assessment reports on organizational structure, and work summaries to extract organizational information, and then converting this information into a structured text sequence. For example, the institutional regulations of the target institution are extracted into data such as institutional information, functional information, staffing, and leadership positions; self-assessment reports are extracted into information on institutional configuration, performance of duties, staffing usage, and leadership allocation; performance of duties in work summaries is extracted into performance information; unstructured data such as lists of powers and responsibilities, internal rules and regulations, business process documents, and public service data are extracted into steps and procedures for performing duties; personnel files, petitions and complaints, news reports, and social media messages are anonymized and cleaned, retaining only key information related to organizational structure, such as the performance of duties, number of personnel, and leadership positions; data from third-party platforms and business systems is obtained, and public service data is standardized into standard data, such as the number of administrative approvals processed, the number of law enforcement cases, and government service satisfaction.

[0026] In step S104 above, organizational entities and their relationships can be extracted from the target organization's organizational structure data. A multi-dimensional profile of the target organization can be created, compiling its current and historical information to support the assessment and analysis of its current operational status and historical issues and solutions. For example, data such as organizational structure information (department names, internal departments, job positions, job descriptions), financial information (budget and final accounts), personnel information (number, age, education, major, job level structure, etc.), performance information (business processes, service recipients, legal basis), and historical issues and solutions can be extracted. Based on the extracted organizational entities and their semantic relationships, the target organization's organizational structure system can be constructed.

[0027] like Figure 3 As shown, after obtaining the organizational structure of the target organization, the scoring rules of each preset dimension in the evaluation standard knowledge base can be used to calculate the evaluation score of the organizational structure in each preset dimension based on the operational feedback data of the target organization.

[0028] For example, in this embodiment, a preset scoring model can be constructed based on the scoring rules for duty performance, mark effectiveness, process efficiency, resource allocation, public satisfaction, and compliance risk. Each preset scoring model is then used to obtain the target organization's evaluation scores in terms of duty performance, mark effectiveness, process efficiency, resource allocation, and compliance risk based on the target organization's operational feedback data.

[0029] Each of the aforementioned preset scoring models can be a neural network model, such as a model based on a recurrent neural network. Taking a preset scoring model for one preset dimension as an example, staff can obtain operational feedback data from multiple reference institutions, and based on the scoring rules for that preset dimension and the operational feedback data, obtain the evaluation score for each reference institution in that preset dimension.

[0030] For example, the current operational status of a target organization can be analyzed based on industry standard data compiled by the organization, as well as the list of powers and responsibilities, the list of duties and tasks for each position in the organization's structure, and business processes. This analysis can identify the target organization's coverage of responsibilities, adequacy of core functions, overstepping / misalignment of responsibilities, process compliance, and achievement of performance results, thereby obtaining an evaluation score for the reference organization in the dimension of responsibility performance.

[0031] Alternatively, based on the positions and job performance information within the organizational structure of the reference organization, we can obtain the average workload per person, the efficiency of staffing resource utilization, the sensitivity of dynamic staffing adjustments, the rationality of age and education structure, the matching degree between professional background and job function, workload saturation, and the achievement of quality and efficiency. Based on the average workload per person, the efficiency of staffing resource utilization, and the sensitivity of dynamic staffing adjustments, we can calculate the staffing-to-workload matching degree of the target organization. Based on the aforementioned rationality of age and education structure and the matching degree between professional background and job function, we can calculate the personnel structure-to-job requirement fit degree of the target organization. Based on the aforementioned workload saturation and the achievement of quality and efficiency, we can obtain the balance between workload and efficiency. Based on the staffing-to-workload matching degree, personnel structure-to-job requirement fit degree, and workload-to-efficiency balance degree of the reference organization, we can calculate the evaluation score of the reference organization in the indicator efficiency dimension.

[0032] Alternatively, based on the administrative approval and enforcement procedures outlined in the power and responsibility list and performance item list of the reference institution, the business processes and actual operational data of the target institution can be compared with those of the administrative approval platform and the administrative enforcement platform. By analyzing the process cycle compliance rate, immediate case completion rate, process compliance rate, and appeal rate, redundant links, bottlenecks, and obstacles to cross-departmental collaboration can be identified, thereby obtaining the target institution's evaluation score in the dimension of process efficiency.

[0033] Alternatively, based on the organizational structure information (job positions, job descriptions), financial information (budget and final accounts, fixed asset information), and personnel information (number of employees in different positions) of the reference organization, assess whether the allocation of internal human resources, budget, equipment, and other resources among departments and projects matches the organization's responsibilities, business priorities, and actual needs. Analyze the resource ratio and its matching degree with core responsibilities, the resource guarantee rate for strategic tasks, the human resource input-output ratio, asset utilization rate, budget execution rate and cost control effectiveness, the matching degree between personnel structure and business needs, and the sensitivity of dynamic resource adjustments to obtain the target organization's evaluation score in the process efficiency dimension.

[0034] Alternatively, based on the reference institutions' petitions, complaints, hotline calls, public opinion, and government service evaluation data, assess the institutions' response to demands and problem-solving quality, business processing efficiency, service attitude and professionalism, and environmental and facility convenience. Through sentiment analysis and thematic modeling, quantitative satisfaction can be obtained to obtain the reference institutions' evaluation scores in the public satisfaction dimension.

[0035] Alternatively, based on the responsibilities of the reference institution, analyze the institution's self-assessment report, relevant meeting minutes, and seminar minutes, systematically review the reference institution's operation to ensure its compliance with laws, regulations, policies, and responsibilities. Assess the compliance, collaborative efficiency, risk management capabilities, and continuous improvement mechanisms of the reference institution's operating mechanism, and identify problems such as low efficiency in the use of staffing resources due to poor collaboration among departments such as organization, human resources, and finance, in order to obtain the reference institution's assessment score in the compliance risk dimension.

[0036] This implementation can construct a training dataset for a preset dimension based on the operational feedback data and evaluation scores of each reference institution. Then, a model based on a recurrent neural network is trained using this training dataset until the accuracy of the neural network model is greater than the set accuracy, so as to obtain a predictive scoring model for the preset dimension. Finally, the operational feedback data of the target institution is input into the predictive scoring model to obtain the evaluation score of the target institution for the preset dimension.

[0037] In step S105 above, taking one preset dimension as an example, assume that the quality levels of this preset dimension include Need for Improvement, Pass, Good, and Excellent. The threshold for determining the level of Need for Improvement includes an upper threshold of 60; the threshold for determining the level of Pass includes an upper threshold of 80 and a lower threshold of 60; the threshold for determining the level of Good includes a lower threshold of 80 and an upper threshold of 90; and the lower threshold for Excellent is 90. If the target organization's standard score in this preset dimension is less than 60, then the target organization's quality level in this preset dimension is Need for Improvement; if the target organization's standard score in this preset dimension is greater than or equal to 60 and less than 80, then the target organization's quality level in this preset dimension is Pass; if the target organization's standard score in this preset dimension is greater than or equal to 80 and less than 90, then the target organization's quality level in this preset dimension is Good; and if the target organization's standard score in this preset dimension is greater than 90, then the target organization's quality level in this preset dimension is Excellent.

[0038] As described above, this embodiment constructs scoring rules for multiple preset dimensions based on industry standard data and evaluation reference data related to organizational structure and staffing, thereby building an evaluation standard knowledge base based on multiple preset dimensions. After collecting multi-source data of the target organization from multiple set data sources, the organizational entity of the target organization is extracted from the multi-source data, and the organizational structure and staffing system of the target organization is constructed. Finally, the evaluation standard knowledge base is used to evaluate the organizational structure and staffing system to obtain the merit level of the organizational structure and staffing system in each preset dimension, thus completing the organizational structure and staffing assessment of the target organization. Since this embodiment collects multi-source data of the target organization from multiple set data sources, the accuracy and reliability of the obtained organizational structure and staffing system and the operational feedback data of the target organization can be guaranteed. Furthermore, the use of the evaluation standard knowledge base can evaluate the organizational structure and staffing system of the target organization from multiple preset dimensions, which can improve the comprehensiveness, accuracy, and reliability of the evaluation of the organizational structure and staffing system, thereby ensuring the accuracy of the organizational structure and staffing assessment of the target organization. Moreover, the technical solution of this embodiment does not require manual construction of evaluation standards and the organizational structure and staffing system of the target organization, thus improving the work efficiency of the organizational structure and staffing assessment of the target organization.

[0039] In some embodiments of the present invention, after obtaining the merit levels of the organizational structure of the target organization in each preset dimension in step S105 above, the method further includes: Based on the evaluation scores of the target organization's organizational structure and staffing system in each preset dimension, a staffing evaluation report for the target organization is generated, and the staffing evaluation results of the target organization are presented visually based on the staffing evaluation report.

[0040] In this embodiment, after obtaining the merit level of the target organization in each preset dimension, the generated organizational structure evaluation result view can intuitively display the evaluation results of the target organization in each preset dimension. The following uses two application scenarios as examples to illustrate the generated organizational structure evaluation result view of the target organization.

[0041] Furthermore, in some embodiments of the present invention, the method for visually presenting the staffing assessment results of the target organization based on the staffing assessment report of the target organization includes: generating a staffing score radar chart, a staffing problem heat map, and / or a staffing resource usage trend analysis chart of the target organization.

[0042] In this embodiment, a radar chart is generated based on the target organization's organizational structure and staffing system, using assessment scores across six preset dimensions: duty performance, staffing efficiency, process efficiency, resource allocation, public satisfaction, and compliance risk. This provides users with a clear view of the target organization's assessment scores in each preset dimension. For example, assuming the target organization's duty performance assessment score is 85, staffing efficiency assessment score is 70, process efficiency assessment score is 40, resource allocation assessment score is 75, job satisfaction assessment score is 90, and compliance risk assessment score is 95, then the generated radar chart of the target organization's scores would look like this: Figure 2 As shown.

[0043] In this embodiment, the organizational structure problems of the target organization can be classified, and a heat map of organizational structure problems of the target organization can be constructed based on the number of problems of each type.

[0044] For example, the staffing resource utilization information obtained in this embodiment includes staffing resource utilization rate and designated job turnover rate, where the designated job turnover rate is the ratio between the number of employees leaving designated jobs and the total number of employees in designated jobs. A line chart of the staffing resource utilization of the generated target organization is shown below. Figure 3 As shown in the example, the horizontal axis of this line chart represents time, and the vertical axis represents percentage. The resource utilization rates of the target organization from January to June were 65%, 72%, 78%, 83%, 88%, and 91%, respectively, and the job utilization threshold was 80%. The set job turnover rates of the target organization from January to June were 2%, 3%, 1%, 5%, 8%, and 12%, respectively, and the job turnover threshold was 10%.

[0045] In the above embodiments, a scoring radar chart and a line chart of resource utilization are used as views of the organizational assessment results of the target organization. It is understood that the above embodiments are exemplary and not restrictive. In other embodiments, a heat map or trend chart can be used as views of the organizational assessment results of the target organization to show users the concentration trends or development trends in specific aspects to assist in decision-making and understanding.

[0046] In some embodiments of the present invention, before collecting data from multiple set data sources in step S103, the method further includes: determining whether the target organization triggers a preset evaluation standard. If so, then proceed with the steps of collecting data from multiple specified data sources.

[0047] In this embodiment, the preset evaluation criteria include: the duration of operation of the target organization is greater than the set evaluation duration threshold, or the interval between the last evaluation and the previous evaluation is greater than the set evaluation duration threshold.

[0048] The technical solution of this embodiment can periodically and continuously evaluate the organizational structure of the target organization, thereby improving the accuracy and reliability of the evaluation of the target organization's organizational structure.

[0049] In some embodiments of the present invention, after obtaining the merit level of the organizational structure system of the target organization in each preset dimension in step S105, the method further includes: weighting and fusing the standard scores of the organizational structure system of the target organization in each preset dimension to obtain the comprehensive evaluation result of the organizational structure of the target organization.

[0050] In this embodiment, it is assumed that the number of preset dimensions is n, where the weight of the i-th preset dimension is... Furthermore, the standard score of the target organization's organizational structure in the i-th preset dimension is... If the overall score of the target organization's organizational structure is P, then... In this embodiment, by weighted summing of the standard scores of the target organization's organizational structure in each preset dimension to obtain the comprehensive evaluation result of the target organization's organizational structure, not only can the accuracy and reliability of the evaluation of the target organization's organizational structure be improved, but the efficiency of the evaluation work can also be improved.

[0051] In some embodiments of the present invention, after constructing the organizational structure system of the target organization based on the organizational entities of the target organization and the relationships between them in step S103, the method further includes: constructing an organizational structure knowledge graph of the target organization based on the organizational structure system of the target organization.

[0052] In this embodiment, a multi-dimensional profile of the organization can be created, aggregating its current and historical operational status to support the assessment and analysis of its current operational status, historical issues, and their resolution. For example, data such as organizational structure information (department names, internal departments, job positions, job descriptions), financial information (budget and final accounts), personnel information (number, age, education, major, job level structure, etc.), performance information (business processes, service recipients, legal basis), and historical issues and their resolution can be extracted. Based on the extracted organizational entities and their semantic relationships, an organizational knowledge graph can be constructed and updated.

[0053] For example, the structured knowledge graph constructed in this embodiment is as follows: Figure 4 As shown, the XX Bureau has an Administrative Approval Section, whose responsibilities include market entity registration and guiding the entire registration process to be electronic. The basis for market entity registration is the Company Law. The positions in the Administrative Approval Section include section chief, deputy section chief, and staff members.

[0054] In this embodiment, after obtaining the organizational structure and staffing system of the target organization, a staffing knowledge graph of the target organization is constructed based on the organizational structure and staffing system to visualize the structure and responsibilities of the target organization's organizational structure and staffing system, so that staff can understand the organizational structure and staffing system of the target organization.

[0055] In some embodiments of the present invention, before using a large language model to perform semantic extraction on the multi-source data of the target organization in step S103 above, the method further includes: preprocessing the multi-source data of the target organization, which includes cleaning, deduplication, format standardization and privacy desensitization.

[0056] In this embodiment, the cleaning process refers to identifying and removing data unrelated to the organizational structure from the target organization's multi-source data; the deduplication process refers to identifying and deleting duplicate data; the format standard conversion process refers to converting the data format of the target organization's multi-source data into a preset format; and the privacy desensitization process refers to removing sensitive personal information from the target organization's multi-source data, such as customers' names, ID card numbers, and contact information.

[0057] In this embodiment, by preprocessing the multi-source data of the target organization, the accuracy, reliability and security of the multi-source data can be improved.

[0058] In some embodiments of the present invention, after obtaining the quality level of the target mechanism in each preset dimension in step S105 above, the method further includes: Determine whether revised evaluation rules for each preset dimension have been received; If received, the evaluation criteria knowledge base will be updated according to the revised content.

[0059] For example, experts can obtain revised evaluation rules for each preset dimension based on experience and prior knowledge. These revisions include deleting existing evaluation rules, adding new evaluation rules, and modifying existing evaluation rules. The revised rules are then sent to a standard knowledge base to modify the evaluation rules for each preset dimension.

[0060] In this embodiment, the accuracy and reliability of the stored scoring rules can be ensured by automatically updating the evaluation criteria knowledge base, thereby improving the accuracy and reliability of the assessment of the target organization's establishment.

[0061] The flowcharts provided in the above embodiments are not intended to indicate that the operations of the methods will be performed in any particular order, or that all operations of the methods are included in all every case. Furthermore, each of the methods described above may include additional operations. Within the scope of the technical concept provided by the methods of these embodiments, additional variations can be made to the above methods.

[0062] It should be understood that in some embodiments, the components may be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods may be implemented using software or firmware stored in memory and executed by a suitable instruction execution system.

[0063] This embodiment also provides an organizational structure assessment system based on a large language model. The system includes a memory 31, a processor 32, and a computer program 11 stored on the memory 31 and running on the processor 32.

[0064] The computer program 11 used to perform the operations of this invention can be assembly instructions, Instruction Set Architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages ​​and procedural programming languages. The computer program 11 can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can connect to the user's computer via any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it can connect to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, to perform aspects of the invention, electronic circuits, including, for example, programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), can execute computer-readable program instructions to personalize the electronic circuits by utilizing state information of computer-readable program instructions.

[0065] Therefore, those skilled in the art should recognize that although numerous exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications conforming to the principles of the present invention can be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Thus, the scope of the present invention should be understood and construed as covering all such other variations or modifications.

Claims

1. A method for evaluating organizational structure based on a large language model, characterized in that, include: Collect industry standard data and evaluation reference data related to institutional establishment, and construct scoring rules with multiple preset dimensions based on the industry standard data and the evaluation reference data; Obtain the level judgment threshold for each of the preset dimensions, and construct an evaluation standard knowledge base based on the scoring rules and the level judgment threshold; Data is collected from multiple set data sources to obtain multi-source data of the target organization, and a large language model is used to extract semantics from the multi-source data to obtain the organizational entities in the target organization and the relationships between them, as well as the operational feedback data of the target organization. The organizational structure of the target organization is constructed based on the organizational entities and the relationships between them, and the organizational structure is evaluated based on the departmental performance data using the scoring rules, so as to obtain the evaluation score of the organizational structure in each of the preset dimensions. Based on each evaluation score and its corresponding level judgment threshold, the merit level of the organizational structure in each preset dimension is obtained.

2. The organizational structure assessment method according to claim 1, characterized in that, The step of constructing scoring rules with multiple preset dimensions based on the industry standard data and the evaluation reference data includes: Based on the industry standard data and the evaluation reference data, scoring rules are obtained for performance of duties, effectiveness of indicators, process efficiency, resource allocation, public satisfaction, and compliance risks.

3. The organizational structure assessment method according to claim 1, characterized in that, After the step of obtaining the merit level of the organization's structure in each preset dimension based on each evaluation score and its corresponding level judgment threshold, the method further includes: A staffing assessment report for the target organization is generated based on each assessment score, and the staffing assessment results of the target organization are visualized based on the staffing assessment report.

4. The organizational structure assessment method according to claim 3, characterized in that, The step of visually presenting the assessment results of the target organization based on the assessment report includes: Based on the assessment report, generate a radar chart of the organization's organizational score, a heat map of organizational issues, and / or a trend analysis chart of organizational resource utilization for the target organization.

5. The organizational structure assessment method according to claim 1, characterized in that, Before the step of collecting data from multiple designated data sources to obtain multi-source data of the target organization, the method further includes: Determine whether the target organization triggers the preset staffing evaluation criteria; If so, then the step of collecting data from multiple set data sources to obtain multi-source data of the target organization is performed.

6. The organizational structure assessment method according to claim 1, characterized in that, After the step of obtaining the merit level of the organization's structure in each preset dimension based on each evaluation score and its corresponding level judgment threshold, the method further includes: Each of the assessment scores is weighted and fused to obtain a comprehensive assessment result for the target institution.

7. The organizational structure assessment method according to claim 1, characterized in that, Following the step of constructing the organizational structure of the target organization based on the organizational entities and their relationships, the method further includes: Construct a staffing knowledge graph of the target organization based on the aforementioned staffing system.

8. The organizational structure assessment method according to claim 1, characterized in that, Before the step of semantic extraction from the multi-source data using a large language model, the method further includes: The multi-source data is preprocessed, and the preprocessing includes at least one of cleaning, deduplication, format standardization, and privacy anonymization.

9. The organizational structure assessment method according to claim 1, characterized in that, After the step of constructing the evaluation standard knowledge base based on the scoring rules and the level judgment threshold, the method further includes: Determine whether any modifications to the evaluation rules have been received; If so, the evaluation criteria knowledge base will be updated according to the modifications.

10. An institutional staffing assessment system based on a large language model, characterized in that, It includes a memory, a processor, and a computer program stored in the memory, the processor executing the computer program to implement the steps of the organizational structure evaluation method based on a large language model as described in any one of claims 1-9.

Images