Management entropy model optimization method and system based on AI construction

Abstract

The invention relates to the technical field of model optimization, and particularly discloses a management entropy model optimization method and system based on AI construction, and the method comprises the steps: carrying out the optimization of a management entropy model based on all enterprise evaluation dimensions of each industry and all evaluation indexes under each enterprise evaluation dimension of each industry; constructing a multi-dimensional evaluation index system for enterprise management entropy evaluation of each industry; determining an optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of the enterprise management entropy evaluation of each industry and an optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry based on an AI algorithm; constructing a management entropy measurement optimization model of each industry based on the optimization dimension weight of each enterprise evaluation dimension of each industry and the optimization index weight of each evaluation index under each enterprise evaluation dimension of each industry; accurate optimization of the management entropy measurement model of each industry is realized, and the measurement precision of the management entropy measurement model on the enterprise management entropy is improved.

Description

Technical Field

[0001] The present invention relates to the field of model optimization technology, and in particular to a management entropy model optimization method and system based on AI. Background Art

[0002] At present, enterprises are facing an increasingly complex business environment. How to accurately assess the health of enterprises has become the focus of attention of enterprise managers and investors. Traditional enterprise management evaluation methods mostly rely on qualitative analysis and expert experience, lack objectivity and quantitative indicators, and cannot fully and timely reflect the overall picture of enterprise management. With the development of artificial intelligence technology, especially the progress of big data analysis and machine learning algorithms, new quantitative tools and methods have been provided for enterprise management. Management entropy, as an emerging enterprise evaluation theory, can effectively explain and evaluate the orderliness of internal management of enterprises. By describing the changes in the orderliness of enterprise management, it can help enterprises identify management problems and improvement directions.

[0003] However, the existing management entropy measurement model still has the problem that the dimension weights of the evaluation dimensions and the indicator weights of the evaluation indicators under the evaluation dimensions are not accurate enough, which will directly affect the measurement accuracy of the management entropy measurement model. For example, the Chinese invention patent with the publication number "CN114881477A" and the patent name "A system and method for evaluating enterprise management entropy" discloses a system and method for evaluating enterprise management entropy, and the method includes: S1, obtaining data that can characterize the operation status of the enterprise, including financial operation status data, innovation and development status data, and personnel management status data; S2, performing data preprocessing on the acquired financial operation status data, innovation and development status data, and personnel management status data of the enterprise to obtain standardized enterprise data; S3, based on the standardized financial operation status data, innovation and development status data, and personnel management status data of the enterprise, constructing a multi-layer evaluation model for enterprise management entropy, and completing the evaluation of enterprise management entropy by using the multi-layer evaluation model, the present invention can solve the problem of low accuracy of the evaluation results of enterprise management entropy in the prior art. However, the patent does not specify the method for determining the dimension weights of the evaluation dimensions and the indicator weights of the evaluation indicators under the evaluation dimensions used in the multi-layer evaluation model of enterprise management entropy, resulting in the inability to guarantee the measurement accuracy of the management entropy measurement model.

[0004] Therefore, the present invention proposes a management entropy model optimization method and system based on AI. Summary of the invention

[0005] The present invention provides a management entropy model optimization method and system based on AI, which realizes the precise optimization of the management entropy measurement model of each industry, thereby improving the measurement accuracy of the management entropy measurement model for the enterprise management entropy.

[0006] The present invention provides a management entropy model optimization method based on AI, comprising: S1: Based on all enterprise evaluation dimensions of each industry and all evaluation indicators under each enterprise evaluation dimension of each industry, a multi-dimensional evaluation indicator system for enterprise management entropy evaluation of each industry is constructed; S2: Based on the AI ​​algorithm, determine the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry; S3: Based on the optimized dimension weights of each enterprise evaluation dimension in each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry, a management entropy measurement optimization model for each industry is constructed.

[0007] Preferably, S2: determining the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry based on the AI ​​algorithm, including: S201: Obtain and summarize a large number of enterprise management entropy measurement instances of all enterprises whose customer satisfaction is not less than a satisfaction threshold as a total set of reference enterprise management entropy measurement instances; S202: classifying the enterprises corresponding to each enterprise management entropy measurement instance in the total set of reference enterprise management entropy measurement instances into industry categories, and obtaining a subset of reference enterprise management entropy measurement instances in each industry; S203: Perform cluster analysis on a subset of reference enterprise management entropy measurement instances of each industry based on the principle of similar entropy values, and obtain the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry.

[0008] Preferably, S203: cluster analysis is performed on a subset of reference enterprise management entropy measurement instances of each industry based on the principle of similar entropy values ​​to obtain the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry, including: Based on the principle of similar entropy values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the first set division result of each set division process, wherein the first set division result of each set division process contains multiple instance clusters, and each instance cluster contains at least multiple enterprise management entropy measurement instances; Calculate the entropy similarity of each instance cluster in the first set partitioning result of each set partitioning process, and take the average of the entropy similarity of all instance clusters in the first set partitioning result of each set partitioning process as the partition satisfaction of each first set partitioning result; Based on the multiple instance clusters contained in the first set partitioning results of the set partitioning process of the maximum partitioning satisfaction of each industry, the optimized dimension weights of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension of each industry are analyzed.

[0009] Preferably, based on the multiple instance clusters contained in the first set partitioning result of the set partitioning process of the maximum partitioning satisfaction of each industry, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are analyzed, including: Based on the principle of similar index values, each instance cluster contained in the first set partitioning result of the set partitioning process of the maximum partitioning satisfaction of each industry is randomly partitioned multiple times to obtain the first cluster partitioning result of each cluster partitioning process, wherein the first cluster partitioning result of each cluster partitioning process contains multiple first sub-instance clusters, and each first sub-instance cluster contains multiple enterprise management entropy measurement instances; Calculate the similarity of all evaluation values ​​of the same evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster as the first evaluation value similarity of each evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster; The mean of the similarity of the first evaluation values ​​of all evaluation indicators under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster is taken as the similarity of the evaluation indicator values ​​of each first sub-instance cluster under the corresponding enterprise evaluation dimension, and the mean of the similarity of the evaluation indicator values ​​of all first sub-instance clusters under the same enterprise evaluation dimension in the first cluster division result of each cluster division process is taken as the division satisfaction of each first cluster division result relative to the corresponding enterprise evaluation dimension; Based on all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry, relative to the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry are analyzed.

[0010] Preferably, in all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry, relative to the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are analyzed, including: The sum of the product of the dimension weight values ​​adopted by all enterprise management entropy measurement instances in each first sub-instance cluster in the first cluster division result of the maximum division satisfaction degree of each industry in all cluster division processes of each instance cluster contained in the first set division result of the maximum division satisfaction degree of each industry and the customer satisfaction degree of the corresponding enterprise management entropy measurement instance is taken as the comprehensive dimension weight of the corresponding first sub-instance cluster relative to the corresponding enterprise evaluation dimension; The average of the customer satisfaction of all enterprise management entropy measurement instances in each first sub-implementation cluster is regarded as the average satisfaction of each first sub-implementation cluster; The sum of the products of the comprehensive dimension weights and the average satisfaction of all the first sub-instance clusters in the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension in all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry is taken as the optimized dimension weight of the corresponding enterprise evaluation dimension of the corresponding industry; Based on the principle of similar dimension evaluation values, a cluster analysis is performed again on the reference enterprise management entropy measurement instance subset of each industry to obtain the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry.

[0011] Preferably, based on the principle of similar dimension evaluation values, a cluster analysis is performed again on the reference enterprise management entropy measurement instance subset of each industry to obtain the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry, including: Based on the principle of similar dimension evaluation values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the second set division result of each set division process for each industry, wherein the second set division result of each set division process contains multiple instance clusters, and each instance cluster contains at least multiple enterprise management entropy measurement instances; Calculate the similarity of the evaluation values ​​of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster in the second set partitioning result of each set partitioning process, and regard it as the similarity of the dimension evaluation value of the corresponding instance cluster relative to the corresponding enterprise evaluation dimension; The average of the similarity of dimension evaluation values ​​of all instance clusters in the second set partitioning results of each set partitioning process relative to the same enterprise evaluation dimension is regarded as the partition satisfaction of each second set partitioning result relative to the corresponding enterprise evaluation dimension; All instance clusters in the second set of partitioning results of each industry with respect to the maximum partitioning satisfaction of each enterprise evaluation dimension are regarded as all reference instance clusters of each enterprise evaluation dimension of the corresponding industry; Based on all reference instance clusters of each enterprise evaluation dimension in each industry, the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry is analyzed.

[0012] Preferably, based on all reference instance clusters of each enterprise evaluation dimension of each industry, the optimization indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry is analyzed, including: Based on the principle of similar index values, cluster analysis is performed on each reference instance cluster of each enterprise evaluation dimension of each industry, and the second cluster division result corresponding to the maximum division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry relative to the corresponding enterprise evaluation dimension is obtained, wherein the second cluster division result includes multiple second sub-instance clusters, and each second sub-instance cluster includes multiple enterprise management entropy measurement instances; In the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry is the maximum relative to the corresponding enterprise evaluation dimension, the sum of the product of the indicator weights adopted by all enterprise management entropy measurement instances of each second sub-instance cluster and the customer satisfaction of the corresponding enterprise management entropy measurement instance is taken as the first indicator weight of the corresponding second sub-instance cluster; The mean of the customer satisfaction of all enterprise management entropy measurement instances of each second sub-instance cluster is taken as the average satisfaction of the corresponding second sub-instance cluster; The sum of the products of the first indicator weights and the average satisfaction of all the second sub-instance clusters in the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry is the maximum relative to the corresponding enterprise evaluation dimension is taken as the second indicator weight of the corresponding reference instance cluster; The sum of the second indicator weights of all reference instance clusters of each enterprise evaluation dimension of each industry and the mean of the average satisfaction of all second sub-instance clusters contained in the corresponding reference instance cluster is taken as the optimized indicator weight of the corresponding evaluation indicator under the corresponding enterprise evaluation dimension of the corresponding industry.

[0013] Preferably, after constructing the management entropy measurement optimization model for each industry, it also includes: Input the valid enterprise data of the enterprise to be analyzed into the management entropy measurement optimization model of the corresponding industry to obtain the management entropy value of the enterprise to be analyzed; The enterprise entropy flow of the enterprise to be analyzed is constructed based on the management entropy values ​​of the enterprise to be analyzed in multiple consecutive years; Construct the industry entropy flow of the industry to which the enterprise to be analyzed belongs; Based on the enterprise entropy flow of the enterprise to be analyzed and the industry entropy flow of the corresponding industry, the internal and external entropy flow analysis results of the enterprise to be analyzed are obtained.

[0014] Preferably, after obtaining the management entropy value of the enterprise to be analyzed, the method further includes: Determine the relative position of the management entropy value of the enterprise to be analyzed within its industry; Based on the relative position of the management entropy value of the enterprise to be analyzed in its industry and the industry performance standard value, the enterprise score value of the enterprise to be analyzed is obtained.

[0015] The present invention provides a management entropy model optimization system based on AI, comprising: The system construction module is used to construct a multi-dimensional evaluation index system for enterprise management entropy evaluation in each industry based on all enterprise evaluation dimensions in each industry and all evaluation indicators under each enterprise evaluation dimension in each industry; The weight optimization module is used to determine the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry based on the AI ​​algorithm; The model optimization module is used to construct a management entropy measurement optimization model for each industry based on the optimized dimension weights of each enterprise evaluation dimension in each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: by constructing a multidimensional evaluation index system for enterprise management entropy evaluation of each industry, multiple evaluation dimensions considered in the enterprise management entropy evaluation of each industry and multiple evaluation indicators under each evaluation dimension are clarified, and based on the AI ​​algorithm, the optimized dimension weights of each enterprise evaluation dimension in the multidimensional evaluation index system for enterprise management entropy evaluation of each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension of each industry are determined, and the management entropy measurement model of each industry is accurately optimized, thereby improving the measurement accuracy of the management entropy measurement model for enterprise management entropy.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0018] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of a management entropy model optimization method based on AI in an embodiment of the present invention; Figure 2 Schematic diagram of a multi-dimensional evaluation index system for enterprise management entropy evaluation in an embodiment of the present invention; Figure 3 is a flowchart of a specific execution method of step S2 in an embodiment of the present invention; Figure 4 This is a schematic diagram showing an example of a dimension weight table for different industries in an embodiment of the present invention; Figure 5 This is a schematic diagram of the internal functional modules of the management entropy model optimization system built based on AI in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0021] Example 1 The present invention provides a management entropy model optimization method based on AI, referring to Figure 1 and 2 ,include: S1: Based on all enterprise evaluation dimensions of each industry and all evaluation indicators under each enterprise evaluation dimension of each industry, a multi-dimensional evaluation indicator system for enterprise management entropy evaluation of each industry is constructed; S2: Based on the AI ​​algorithm, determine the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry; S3: Based on the optimized dimension weights of each enterprise evaluation dimension in each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry, a management entropy measurement optimization model for each industry is constructed.

[0022] In this embodiment, the enterprise evaluation dimensions are dimensions for evaluating the effective energy of the enterprise, such as profitability, operational capability, growth capability, innovation capability, cost management capability, financial management capability, debt repayment capability, investment and financing capability, asset management capability, and human efficiency management capability.

[0023] In this embodiment, all evaluation indicators under each enterprise evaluation dimension are evaluation indicators used to obtain the evaluation value of a single enterprise evaluation dimension. For example, all evaluation indicators of operating capability include operating profit margin, cost profit margin, surplus cash protection ratio, return on total assets, return on net assets and return on capital.

[0024] In this embodiment, the multi-dimensional evaluation index system for enterprise management entropy evaluation of the industry is as follows: Figure 2 As shown, the management capabilities of enterprises are evaluated from 10 dimensions, namely profitability, operational capability, growth capability, innovation capability, cost management capability, financial management capability, debt repayment capability, investment and financing capability, asset management capability, and human efficiency management capability. At least three representative evaluation indicators are selected for each dimension, thus constructing a comprehensive management entropy evaluation indicator system.

[0025] In this embodiment, the AI ​​algorithm here mainly refers to big data analysis technology.

[0026] In this embodiment, the optimization dimension weight represents the proportion of the evaluation value of the evaluation dimension of a single enterprise in a single industry in the entropy value of a single enterprise in the corresponding industry.

[0027] In this embodiment, the optimization indicator weight represents the proportion of the indicator value of a single evaluation indicator in the evaluation value of the corresponding enterprise evaluation dimension.

[0028] In this embodiment, based on the optimized dimension weight of each enterprise evaluation dimension of each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry, a management entropy measurement optimization model for each industry is constructed, namely: The calculation method disclosed in the prior art for each evaluation indicator under each enterprise evaluation dimension in each industry is used as the first step of the calculation process; Based on the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry, the index values ​​of each evaluation index under each enterprise evaluation dimension of each industry obtained in the first step of the calculation process are summed up according to the weights as the second step of the calculation process; Based on the optimized dimension weights of each enterprise evaluation dimension in each industry, the evaluation values ​​of each enterprise evaluation dimension in each industry obtained in the second step of the calculation process are summed up according to the weights to obtain the management entropy value of the enterprise; The calculation logic based on the above multiple calculation processes is used to build a management entropy measurement optimization model for each industry.

[0029] The beneficial effects of the above technology are: by constructing a multi-dimensional evaluation index system for enterprise management entropy evaluation in each industry, the multiple evaluation dimensions considered in the enterprise management entropy evaluation of each industry and the multiple evaluation indicators under each evaluation dimension are clarified, and based on the AI ​​algorithm, the optimized dimension weights of each enterprise evaluation dimension in the multi-dimensional evaluation index system for enterprise management entropy evaluation of each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry are determined, and the management entropy measurement model of each industry is accurately optimized, thereby improving the measurement accuracy of the management entropy measurement model for enterprise management entropy.

[0030] Example 2 On the basis of Example 1, S2: Based on the AI ​​algorithm, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are determined, referring to Figure 3 and 4 ,include: S201: Obtain and summarize a large number of enterprise management entropy measurement instances of all enterprises whose customer satisfaction is not less than a satisfaction threshold as a total set of reference enterprise management entropy measurement instances; S202: classifying the enterprises corresponding to each enterprise management entropy measurement instance in the total set of reference enterprise management entropy measurement instances into industry categories, and obtaining a subset of reference enterprise management entropy measurement instances in each industry; S203: Perform cluster analysis on a subset of reference enterprise management entropy measurement instances of each industry based on the principle of similar entropy values, and obtain the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry.

[0031] In this embodiment, customer satisfaction indicates the degree of satisfaction of the customer with the management entropy value of the corresponding enterprise measured in the enterprise management entropy measurement instance, and to a certain extent can indicate the accuracy of the management entropy value of the corresponding enterprise measured in the enterprise management entropy measurement instance.

[0032] In this embodiment, the satisfaction threshold is a preset value that cannot be less than the satisfaction of the enterprise management entropy measurement instance included in the total set of reference enterprise management entropy measurement instances.

[0033] In this embodiment, the enterprise management entropy measurement instance records the calculation process of calculating the management entropy value of a single enterprise, specifically including the indicator weight of each evaluation indicator used when calculating the evaluation value of each enterprise evaluation dimension and the dimension weight of each enterprise evaluation dimension used when calculating the enterprise management entropy value.

[0034] In this embodiment, the industry classification is performed on the enterprise corresponding to each enterprise management entropy measurement instance in the total set of reference enterprise management entropy measurement instances, and a clustering model is used to classify all industries into multiple industry categories, such as Figure 4 Agriculture, forestry, animal husbandry and fishery, energy and raw material production, construction industry, etc.

[0035] In this embodiment, all enterprises corresponding to the enterprise management entropy measurement instances included in the reference enterprise management entropy measurement instance subset of each industry belong to the industry.

[0036] The beneficial effects of the above technology are: by acquiring the total set of reference enterprise management entropy measurement instances and classifying them by industry, a subset of reference enterprise management entropy measurement instances in each industry is obtained, and cluster analysis is performed on the subset of reference enterprise management entropy measurement instances in each industry based on the principle of similar entropy values, so as to achieve accurate determination of the optimized dimension weights of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy assessment for each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry.

[0037] Example 3 On the basis of Example 2, S203: cluster analysis is performed on a subset of reference enterprise management entropy measurement instances of each industry based on the principle of similar entropy values, and the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are obtained, including: Based on the principle of similar entropy values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the first set division result of each set division process, wherein the first set division result of each set division process contains multiple instance clusters, and each instance cluster contains at least multiple enterprise management entropy measurement instances; Calculate the entropy similarity of each instance cluster in the first set partitioning result of each set partitioning process, and take the average of the entropy similarity of all instance clusters in the first set partitioning result of each set partitioning process as the partition satisfaction of each first set partitioning result; Based on the multiple instance clusters contained in the first set partitioning results of the set partitioning process of the maximum partitioning satisfaction of each industry, the optimized dimension weights of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension of each industry are analyzed.

[0038] In this embodiment, based on the principle of similar entropy values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the first set division result of each set division process, which means that the management entropy values ​​measured in all enterprise management entropy measurement instances contained in each instance cluster contained in each set division process are relatively similar; Among them, all enterprise management entropy measurement instances contained in all instance clusters contained in the first set of partitioning results are all enterprise management entropy measurement instances contained in the reference enterprise management entropy measurement instance subset of the corresponding industry; And there are no repeated enterprise management entropy measurement instances among all instance clusters included in each first set of partitioning results.

[0039] In this embodiment, the entropy similarity of each instance cluster in the first set partitioning result of each set partitioning process is calculated, including: The absolute value of the difference between the management entropy values ​​measured by all enterprise management entropy measurement instances in each instance cluster in the first set partitioning result of each set partitioning process and the mean of the management entropy values ​​measured by all enterprise management entropy measurement instances in the corresponding instance cluster is averaged to obtain the entropy value separation degree of the corresponding instance cluster; The difference between 1 and the entropy value of each instance cluster is regarded as the entropy similarity of each instance cluster.

[0040] In this embodiment, the division satisfaction of the first set division result indicates the degree to which the first set division result satisfies the principle of similar entropy values.

[0041] The beneficial effects of the above technology are: by using the principle of similar entropy values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times, and the division satisfaction of the first set division result of each set division process is calculated. Based on the first set division result corresponding to the maximum division satisfaction, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry are accurately analyzed.

[0042] Example 4 On the basis of Example 3, based on the multiple instance clusters contained in the first set partitioning result of the set partitioning process of the maximum partitioning satisfaction of each industry, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are analyzed, including: Based on the principle of similar index values, each instance cluster contained in the first set partitioning result of the set partitioning process of the maximum partitioning satisfaction of each industry is randomly partitioned multiple times to obtain the first cluster partitioning result of each cluster partitioning process, wherein the first cluster partitioning result of each cluster partitioning process contains multiple first sub-instance clusters, and each first sub-instance cluster contains multiple enterprise management entropy measurement instances; Calculate the similarity of all evaluation values ​​of the same evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster as the first evaluation value similarity of each evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster; The mean of the similarity of the first evaluation values ​​of all evaluation indicators under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster is taken as the similarity of the evaluation indicator values ​​of each first sub-instance cluster under the corresponding enterprise evaluation dimension, and the mean of the similarity of the evaluation indicator values ​​of all first sub-instance clusters under the same enterprise evaluation dimension in the first cluster division result of each cluster division process is taken as the division satisfaction of each first cluster division result relative to the corresponding enterprise evaluation dimension; Based on all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry, relative to the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry are analyzed.

[0043] In this embodiment, based on the principle of similar index values, each instance cluster contained in the first set partitioning result of the set partitioning process of the maximum partition satisfaction of each industry is randomly partitioned multiple times to obtain the first cluster partitioning result of each cluster partitioning process, which means that the evaluation values ​​of the same evaluation index measured in all enterprise management entropy measurement instances contained in each first sub-instance cluster contained in each cluster partitioning process are relatively similar; Among them, all enterprise management entropy measurement instances contained in all first sub-instance clusters contained in the first cluster division result are all enterprise management entropy measurement instances contained in the corresponding instance cluster; Furthermore, there are no repeated enterprise management entropy measurement instances among all first sub-instance clusters included in each first cluster division result.

[0044] In this embodiment, the similarity of all evaluation values ​​of the same evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster is calculated, including: The absolute value of the difference between all evaluation values ​​of the same evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster and the mean of all evaluation values ​​of the same evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in the corresponding first sub-instance cluster is averaged to obtain the degree of separation of all evaluation values ​​of each first sub-instance cluster; The difference between 1 and the degree of separation of all evaluation values ​​of each first sub-instance cluster is regarded as the degree of similarity of all evaluation values ​​of each first sub-instance cluster.

[0045] In this embodiment, the division satisfaction of each first cluster division result relative to the corresponding enterprise evaluation dimension represents the degree to which each first cluster division result satisfies the principle of similar index values ​​relative to the corresponding enterprise evaluation dimension.

[0046] The beneficial effects of the above technology are: by using the principle of similar indicator values, each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry is arbitrarily divided multiple times, and the division satisfaction of each first cluster division result relative to the corresponding enterprise evaluation dimension is calculated. Based on the first cluster division result corresponding to the maximum division satisfaction, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry are accurately analyzed.

[0047] Embodiment 5: On the basis of Example 4, in all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry, relative to the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are analyzed, including: The sum of the product of the dimension weight values ​​adopted by all enterprise management entropy measurement instances in each first sub-instance cluster in the first cluster division result of the maximum division satisfaction degree of each industry in all cluster division processes of each instance cluster contained in the first set division result of the maximum division satisfaction degree of each industry and the customer satisfaction degree of the corresponding enterprise management entropy measurement instance is taken as the comprehensive dimension weight of the corresponding first sub-instance cluster relative to the corresponding enterprise evaluation dimension; The average of the customer satisfaction of all enterprise management entropy measurement instances in each first sub-implementation cluster is regarded as the average satisfaction of each first sub-implementation cluster; The sum of the products of the comprehensive dimension weights and the average satisfaction of all the first sub-instance clusters in the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension in all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry is taken as the optimized dimension weight of the corresponding enterprise evaluation dimension of the corresponding industry; Based on the principle of similar dimension evaluation values, a cluster analysis is performed again on the reference enterprise management entropy measurement instance subset of each industry to obtain the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry.

[0048] The beneficial effects of the above technology are: on the basis of the first cluster division result of the maximum division satisfaction of the first set division result of the maximum division satisfaction, the dimension weight values ​​adopted by all enterprise management entropy measurement instances and the customer satisfaction of the corresponding enterprise management entropy measurement instances are used to perform weighted summation layer by layer, and the big data technology is reasonably used to accurately analyze the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry, and further, based on the principle of similar dimension evaluation values, the reference enterprise management entropy measurement instance subset of each industry is clustered again to accurately analyze the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry.

[0049] Example 6 On the basis of Example 5, based on the principle of similar dimension evaluation values, a cluster analysis is performed again on the reference enterprise management entropy measurement instance subset of each industry to obtain the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry, including: Based on the principle of similar dimension evaluation values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the second set division result of each set division process for each industry, wherein the second set division result of each set division process contains multiple instance clusters, and each instance cluster contains at least multiple enterprise management entropy measurement instances; Calculate the similarity of the evaluation values ​​of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster in the second set partitioning result of each set partitioning process, and regard it as the similarity of the dimension evaluation value of the corresponding instance cluster relative to the corresponding enterprise evaluation dimension; The average of the similarity of dimension evaluation values ​​of all instance clusters in the second set partitioning results of each set partitioning process relative to the same enterprise evaluation dimension is regarded as the partition satisfaction of each second set partitioning result relative to the corresponding enterprise evaluation dimension; All instance clusters in the second set of partitioning results of each industry with respect to the maximum partitioning satisfaction of each enterprise evaluation dimension are regarded as all reference instance clusters of each enterprise evaluation dimension of the corresponding industry; Based on all reference instance clusters of each enterprise evaluation dimension in each industry, the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry is analyzed.

[0050] In this embodiment, based on the principle of similar dimension evaluation values, the reference enterprise management entropy measurement instance subset of each industry is arbitrarily divided multiple times to obtain the second set division result of each set division process for each industry, which means that the evaluation values ​​of the same enterprise evaluation dimension in all enterprise management entropy measurement instances contained in each instance cluster contained in each second set division result are relatively similar; Among them, all enterprise management entropy measurement instances contained in all instance clusters contained in each second set division result are all enterprise management entropy measurement instances contained in the reference enterprise management entropy measurement instance subset of the corresponding industry; And there are no repeated enterprise management entropy measurement instances among all instance clusters included in each second set partitioning result.

[0051] In this embodiment, the similarity of evaluation values ​​of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster in the second set division result of each set division process is calculated, including: The absolute value of the difference between all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster in the second set division result of each set division process and the mean of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in the corresponding instance cluster is averaged to obtain the evaluation value deviation of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in the corresponding instance cluster; The difference between 1 and the evaluation value deviation of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster is regarded as the evaluation value similarity of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster.

[0052] In this embodiment, the division satisfaction of each second set division result relative to the corresponding enterprise evaluation dimension represents the degree to which each second set division result satisfies the principle of similar dimension evaluation values ​​relative to the corresponding enterprise evaluation dimension.

[0053] The beneficial effects of the above technology are: by using the principle of similar dimensional evaluation values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times, and the division satisfaction of each second set of division results relative to the corresponding enterprise evaluation dimension is calculated, and based on the second set division results of each industry with the maximum division satisfaction relative to each enterprise evaluation dimension, all reference instance clusters of each enterprise evaluation dimension of each industry are determined, and finally, based on all reference instance clusters of each enterprise evaluation dimension of each industry, the optimization indicator weights of each evaluation indicator under each enterprise evaluation dimension of each industry are accurately analyzed.

[0054] Embodiment 7: On the basis of Example 6, based on all reference instance clusters of each enterprise evaluation dimension of each industry, the optimization index weight of each evaluation index under each enterprise evaluation dimension of each industry is analyzed, including: Based on the principle of similar index values, cluster analysis is performed on each reference instance cluster of each enterprise evaluation dimension in each industry to obtain the second cluster division result corresponding to the maximum division satisfaction of each reference instance cluster of each enterprise evaluation dimension in each industry relative to the corresponding enterprise evaluation dimension, including: Based on the principle of similar index values, each reference instance cluster of each enterprise evaluation dimension in each industry is arbitrarily divided multiple times to obtain the second cluster division result of each cluster division process, wherein the second cluster division result of each cluster division process contains multiple second sub-instance clusters, and each second sub-instance cluster contains multiple enterprise management entropy measurement instances; Calculate the similarity of all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster as the second evaluation value similarity of each evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster; The average of the similarity of the second evaluation values ​​of all evaluation indicators under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster is taken as the similarity of the evaluation indicator values ​​of each second sub-instance cluster under the corresponding enterprise evaluation dimension, and the average of the similarity of the evaluation indicator values ​​of all second sub-instance clusters under the corresponding enterprise evaluation dimension in the second cluster division result of each cluster division process is taken as the division satisfaction of each second cluster division result relative to the corresponding enterprise evaluation dimension; In all cluster division processes of each reference instance cluster of each enterprise evaluation dimension of each industry, the second cluster division result when the division satisfaction with respect to the corresponding enterprise evaluation dimension is the largest is selected; The second cluster division result includes a plurality of second sub-instance clusters, and each second sub-instance cluster includes a plurality of enterprise management entropy measurement instances; In the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry is the maximum relative to the corresponding enterprise evaluation dimension, the sum of the product of the indicator weights adopted by all enterprise management entropy measurement instances of each second sub-instance cluster and the customer satisfaction of the corresponding enterprise management entropy measurement instance is taken as the first indicator weight of the corresponding second sub-instance cluster; The mean of the customer satisfaction of all enterprise management entropy measurement instances of each second sub-instance cluster is taken as the average satisfaction of the corresponding second sub-instance cluster; The sum of the products of the first indicator weights and the average satisfaction of all the second sub-instance clusters in the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry is the maximum relative to the corresponding enterprise evaluation dimension is taken as the second indicator weight of the corresponding reference instance cluster; The sum of the second indicator weights of all reference instance clusters of each enterprise evaluation dimension of each industry and the mean of the average satisfaction of all second sub-instance clusters contained in the corresponding reference instance cluster is taken as the optimized indicator weight of the corresponding evaluation indicator under the corresponding enterprise evaluation dimension of the corresponding industry.

[0055] In this embodiment, based on the principle of similar indicator values, each reference instance cluster of each enterprise evaluation dimension in each industry is arbitrarily divided multiple times to obtain a second cluster division result for each cluster division process, which means that all indicator values ​​of the same evaluation indicator measured in all enterprise management entropy measurement instances contained in all second sub-instance clusters contained in each second cluster division result are relatively similar.

[0056] Among them, all the enterprise management entropy measurement instances contained in all the second sub-instance clusters contained in each second cluster division result are all the enterprise management entropy measurement instances contained in the corresponding reference instance cluster; Furthermore, there are no repeated enterprise management entropy measurement instances among all the second sub-instance clusters included in each second cluster division result.

[0057] In this embodiment, the similarity of all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster is calculated, including: The absolute value of the difference between all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster and the mean of all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in the corresponding sub-instance cluster is averaged to obtain the degree of separation of all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in the corresponding second sub-instance cluster; The difference between 1 and the degree of separation of all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster is regarded as the degree of similarity of all evaluation values ​​of the same evaluation indicator under the corresponding enterprise evaluation dimension in all enterprise management entropy measurement instances in each second sub-instance cluster.

[0058] In this embodiment, the degree of satisfaction of each second cluster division result relative to the division of the corresponding enterprise evaluation dimension represents the degree to which the second cluster division result satisfies the principle of similar index values.

[0059] The beneficial effects of the above technology are: by using the principle of similar indicator values, each reference instance cluster of each enterprise evaluation dimension in each industry is arbitrarily divided multiple times, and the division satisfaction of each second cluster division result relative to the corresponding enterprise evaluation dimension is calculated, and based on the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension in each industry is the largest relative to the corresponding enterprise evaluation dimension and the customer satisfaction of the corresponding enterprise management entropy measurement instance, the indicator weights adopted by all enterprise management entropy measurement instances of all second sub-instance clusters are summed layer by layer according to the weights, so as to obtain a reasonable and accurate optimized indicator weight for each evaluation indicator under each enterprise evaluation dimension in each industry.

[0060] Embodiment 8: On the basis of Example 1, after constructing the management entropy measurement optimization model for each industry, it also includes: Input the valid enterprise data of the enterprise to be analyzed into the management entropy measurement optimization model of the corresponding industry to obtain the management entropy value of the enterprise to be analyzed; The enterprise entropy flow of the enterprise to be analyzed is constructed based on the management entropy values ​​of the enterprise to be analyzed in multiple consecutive years; Construct the industry entropy flow of the industry to which the enterprise to be analyzed belongs; Based on the enterprise entropy flow of the enterprise to be analyzed and the industry entropy flow of the corresponding industry, the internal and external entropy flow analysis results of the enterprise to be analyzed are obtained.

[0061] In this embodiment, the enterprise to be analyzed represents an enterprise whose management entropy value needs to be measured using a management entropy measurement optimization model.

[0062] In this embodiment, the effective enterprise data is the enterprise data required to determine the management entropy value of the enterprise to be analyzed by using the management entropy measurement optimization model corresponding to the industry to which it belongs.

[0063] In this embodiment, the management entropy value is a function that describes the business management status of an enterprise and is used to measure the degree of disorder (chaos) or effective energy of the management system. The larger the management entropy value, the more chaotic the enterprise management system is and the lower the effective energy is.

[0064] In this embodiment, the enterprise entropy flow of the enterprise to be analyzed is constructed based on the management entropy values ​​of the enterprise to be analyzed in multiple consecutive years, including: The management entropy values ​​of the enterprise to be analyzed in multiple consecutive years are represented by a bar graph with the horizontal axis being the year and the vertical axis being the management entropy value, so as to obtain a schematic diagram of the enterprise entropy flow of the enterprise to be analyzed.

[0065] In this embodiment, the industry entropy flow of the industry to which the enterprise to be analyzed belongs is constructed, including: Add up the management entropy values ​​of all enterprises in the industry to which the enterprise to be analyzed belongs in the same year, and obtain the industry entropy value of the industry to which the enterprise to be analyzed belongs in each year; The management entropy values ​​of the enterprise to be analyzed in multiple consecutive years are represented by a bar graph with the year as the horizontal axis and the management entropy value as the vertical axis, so as to obtain the industry entropy flow of the industry to which the enterprise to be analyzed belongs.

[0066] In this embodiment, based on the enterprise entropy flow of the enterprise to be analyzed and the industry entropy flow of the corresponding industry, the internal and external entropy flow analysis results of the enterprise to be analyzed are obtained, specifically: According to the entropy increase and decrease of industry entropy flow and enterprise entropy flow, it reflects the decrease or improvement of enterprise management orderliness; For example, when the enterprise entropy flow of the enterprise to be analyzed shows that the management entropy value of the enterprise to be analyzed decreases year by year within a known year, and the industry entropy flow of the industry to which the enterprise to be analyzed belongs shows that the management entropy value of the industry to which the enterprise to be analyzed belongs increases year by year within a known year, it means that the enterprise to be analyzed has overcome the entropy increase brought about by the external environment, grown in adversity, and performed excellently, which is the result of the internal and external entropy flow analysis of the enterprise to be analyzed.

[0067] Embodiment 9: On the basis of Example 2, after obtaining the management entropy value of the enterprise to be analyzed, the method further includes: Determine the relative position of the management entropy value of the enterprise to be analyzed within its industry; Based on the relative position of the management entropy value of the enterprise to be analyzed in its industry and the industry performance standard value, the enterprise score value of the enterprise to be analyzed is obtained.

[0068] In this embodiment, the relative position of the management entropy value of the enterprise to be analyzed in the industry to which it belongs is determined as follows: The management entropy values ​​of all enterprises in the industry to which the enterprise to be analyzed belongs are sorted to obtain the enterprise ranking result in the industry described by the enterprise to be analyzed, and the ratio of the ranking ordinal number of the enterprise to be analyzed in the enterprise ranking result and the total number of all enterprises in the corresponding industry is regarded as the relative position of the management entropy value of the enterprise to be analyzed in the industry to which it belongs.

[0069] In this embodiment, based on the relative position of the management entropy value of the enterprise to be analyzed in the industry to which it belongs and the industry performance standard value, the enterprise score value of the enterprise to be analyzed is obtained, including: The product of the relative position of the management entropy value of the enterprise to be analyzed in the industry and the industry performance standard value is used as the enterprise score value of the enterprise to be analyzed; Among them, the industry performance standard value represents the maximum score that all enterprises in the industry to which the enterprise to be analyzed belongs can achieve; The enterprise score of the enterprise to be analyzed is the value obtained after a reasonable evaluation of the management efficiency of the enterprise to be analyzed, taking into account the entropy values ​​of all enterprises in the industry to which the enterprise to be analyzed belongs.

[0070] The beneficial effect of the above technology is: based on the relative position of the management entropy value of the enterprise to be analyzed in the industry to which it belongs and the industry performance standard value, a reasonable evaluation of the management efficiency of the enterprise to be analyzed can be achieved under the premise of considering the entropy values ​​of all enterprises in the industry to which the enterprise to be analyzed belongs.

[0071] Embodiment 10: The present invention provides a management entropy model optimization system based on AI, referring to Figure 5 ,include: The system construction module is used to construct a multi-dimensional evaluation index system for enterprise management entropy evaluation in each industry based on all enterprise evaluation dimensions in each industry and all evaluation indicators under each enterprise evaluation dimension in each industry; The weight optimization module is used to determine the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry based on the AI ​​algorithm; The model optimization module is used to construct a management entropy measurement optimization model for each industry based on the optimized dimension weights of each enterprise evaluation dimension in each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry.

[0072] The beneficial effects of the above technology are: by constructing a multi-dimensional evaluation index system for enterprise management entropy evaluation in each industry, the multiple evaluation dimensions considered in the enterprise management entropy evaluation of each industry and the multiple evaluation indicators under each evaluation dimension are clarified, and based on the AI ​​algorithm, the optimized dimension weights of each enterprise evaluation dimension in the multi-dimensional evaluation index system for enterprise management entropy evaluation of each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry are determined, and the management entropy measurement model of each industry is accurately optimized, thereby improving the measurement accuracy of the management entropy measurement model for enterprise management entropy.

[0073] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0074] The beneficial effects of the above technology are: based on the effective enterprise data of the enterprise to be analyzed and the management entropy measurement optimization model of the corresponding industry, accurate measurement of the management entropy value of the enterprise to be analyzed is achieved, and the enterprise entropy flow of the enterprise to be analyzed is generated. Combined with the industry entropy flow of the industry to which the enterprise to be analyzed belongs, the analysis of the management efficiency of the enterprise to be analyzed is achieved while considering the overall industry environment.

Claims

1. The management entropy model optimization method based on AI is characterized by: include: S1: Based on all enterprise evaluation dimensions of each industry and all evaluation indicators under each enterprise evaluation dimension of each industry, a multi-dimensional evaluation indicator system for enterprise management entropy evaluation of each industry is constructed; S2: Based on the AI ​​algorithm, determine the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry; S3: Based on the optimized dimension weights of each enterprise evaluation dimension in each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry, a management entropy measurement optimization model for each industry is constructed.

2. The management entropy model optimization method based on AI construction according to claim 1 is characterized in that: S2: Based on the AI ​​algorithm, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry are determined, including: S201: Obtain and summarize a large number of enterprise management entropy measurement instances of all enterprises whose customer satisfaction is not less than a satisfaction threshold as a total set of reference enterprise management entropy measurement instances; S202: classifying the enterprises corresponding to each enterprise management entropy measurement instance in the total set of reference enterprise management entropy measurement instances into industry categories, and obtaining a subset of reference enterprise management entropy measurement instances in each industry; S203: Perform cluster analysis on a subset of reference enterprise management entropy measurement instances of each industry based on the principle of similar entropy values, and obtain the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry.

3. The management entropy model optimization method based on AI construction according to claim 2 is characterized in that: S203: cluster analysis is performed on the reference enterprise management entropy measurement instance subsets of each industry based on the principle of similar entropy values ​​to obtain the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry, including: Based on the principle of similar entropy values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the first set division result of each set division process, wherein the first set division result of each set division process contains multiple instance clusters, and each instance cluster contains at least multiple enterprise management entropy measurement instances; Calculate the entropy similarity of each instance cluster in the first set partitioning result of each set partitioning process, and take the average of the entropy similarity of all instance clusters in the first set partitioning result of each set partitioning process as the partition satisfaction of each first set partitioning result; Based on the multiple instance clusters contained in the first set partitioning results of the set partitioning process of the maximum partitioning satisfaction of each industry, the optimized dimension weights of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension of each industry are analyzed.

4. The management entropy model optimization method based on AI construction according to claim 3 is characterized in that: Based on the multiple instance clusters contained in the first set partitioning result of the set partitioning process of the maximum partitioning satisfaction of each industry, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are analyzed, including: Based on the principle of similar index values, each instance cluster contained in the first set partitioning result of the set partitioning process of the maximum partitioning satisfaction of each industry is randomly partitioned multiple times to obtain the first cluster partitioning result of each cluster partitioning process, wherein the first cluster partitioning result of each cluster partitioning process contains multiple first sub-instance clusters, and each first sub-instance cluster contains multiple enterprise management entropy measurement instances; Calculate the similarity of all evaluation values ​​of the same evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster as the first evaluation value similarity of each evaluation indicator under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster; The mean of the similarity of the first evaluation values ​​of all evaluation indicators under each enterprise evaluation dimension in all enterprise management entropy measurement instances in each first sub-instance cluster is taken as the similarity of the evaluation indicator values ​​of each first sub-instance cluster under the corresponding enterprise evaluation dimension, and the mean of the similarity of the evaluation indicator values ​​of all first sub-instance clusters under the same enterprise evaluation dimension in the first cluster division result of each cluster division process is taken as the division satisfaction of each first cluster division result relative to the corresponding enterprise evaluation dimension; Based on all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry, relative to the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension of each industry are analyzed.

5. The management entropy model optimization method based on AI construction according to claim 4 is characterized in that: Based on the first set partitioning result of the maximum partitioning satisfaction of each industry, in all cluster partitioning processes of each instance cluster, relative to the first cluster partitioning result of the maximum partitioning satisfaction of each enterprise evaluation dimension, the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation of each industry and the optimized index weight of each evaluation index under each enterprise evaluation dimension of each industry are analyzed, including: The sum of the product of the dimension weight values ​​adopted by all enterprise management entropy measurement instances in each first sub-instance cluster in the first cluster division result of the maximum division satisfaction degree of each industry in all cluster division processes of each instance cluster contained in the first set division result of the maximum division satisfaction degree of each industry and the customer satisfaction degree of the corresponding enterprise management entropy measurement instance is taken as the comprehensive dimension weight of the corresponding first sub-instance cluster relative to the corresponding enterprise evaluation dimension; The average of the customer satisfaction of all enterprise management entropy measurement instances in each first sub-implementation cluster is regarded as the average satisfaction of each first sub-implementation cluster; The sum of the products of the comprehensive dimension weights and the average satisfaction of all the first sub-instance clusters in the first cluster division result of the maximum division satisfaction of each enterprise evaluation dimension in all cluster division processes of each instance cluster contained in the first set division result of the set division process of the maximum division satisfaction of each industry is taken as the optimized dimension weight of the corresponding enterprise evaluation dimension of the corresponding industry; Based on the principle of similar dimension evaluation values, a cluster analysis is performed again on the reference enterprise management entropy measurement instance subset of each industry to obtain the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry.

6. The management entropy model optimization method based on AI construction according to claim 5 is characterized in that: Based on the principle of similar dimension evaluation values, a cluster analysis is performed again on the reference enterprise management entropy measurement instance subset of each industry to obtain the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry, including: Based on the principle of similar dimension evaluation values, the reference enterprise management entropy measurement instance subsets of each industry are arbitrarily divided multiple times to obtain the second set division result of each set division process for each industry, wherein the second set division result of each set division process contains multiple instance clusters, and each instance cluster contains at least multiple enterprise management entropy measurement instances; Calculate the similarity of the evaluation values ​​of all dimension evaluation values ​​of the same enterprise evaluation dimension of all enterprise management entropy measurement instances in each instance cluster in the second set partitioning result of each set partitioning process, and regard it as the similarity of the dimension evaluation value of the corresponding instance cluster relative to the corresponding enterprise evaluation dimension; The average of the similarity of dimension evaluation values ​​of all instance clusters in the second set partitioning results of each set partitioning process relative to the same enterprise evaluation dimension is regarded as the partition satisfaction of each second set partitioning result relative to the corresponding enterprise evaluation dimension; All instance clusters in the second set of partitioning results of each industry with respect to the maximum partitioning satisfaction of each enterprise evaluation dimension are regarded as all reference instance clusters of each enterprise evaluation dimension of the corresponding industry; Based on all reference instance clusters of each enterprise evaluation dimension in each industry, the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry is analyzed.

7. The management entropy model optimization method based on AI construction according to claim 6 is characterized in that: Based on all reference instance clusters of each enterprise evaluation dimension in each industry, the optimization indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry are analyzed, including: Based on the principle of similar index values, cluster analysis is performed on each reference instance cluster of each enterprise evaluation dimension of each industry, and the second cluster division result corresponding to the maximum division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry relative to the corresponding enterprise evaluation dimension is obtained, wherein the second cluster division result includes multiple second sub-instance clusters, and each second sub-instance cluster includes multiple enterprise management entropy measurement instances; In the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry is the maximum relative to the corresponding enterprise evaluation dimension, the sum of the product of the indicator weights adopted by all enterprise management entropy measurement instances of each second sub-instance cluster and the customer satisfaction of the corresponding enterprise management entropy measurement instance is taken as the first indicator weight of the corresponding second sub-instance cluster; The mean of the customer satisfaction of all enterprise management entropy measurement instances of each second sub-instance cluster is taken as the average satisfaction of the corresponding second sub-instance cluster; The sum of the products of the first indicator weights and the average satisfaction of all the second sub-instance clusters in the second cluster division result when the division satisfaction of each reference instance cluster of each enterprise evaluation dimension of each industry is the maximum relative to the corresponding enterprise evaluation dimension is taken as the second indicator weight of the corresponding reference instance cluster; The sum of the second indicator weights of all reference instance clusters of each enterprise evaluation dimension of each industry and the mean of the average satisfaction of all second sub-instance clusters contained in the corresponding reference instance cluster is taken as the optimized indicator weight of the corresponding evaluation indicator under the corresponding enterprise evaluation dimension of the corresponding industry.

8. The management entropy model optimization method based on AI construction according to claim 1 is characterized in that: After building the management entropy measurement optimization model for each industry, it also includes: Input the valid enterprise data of the enterprise to be analyzed into the management entropy measurement optimization model of the corresponding industry to obtain the management entropy value of the enterprise to be analyzed; The enterprise entropy flow of the enterprise to be analyzed is constructed based on the management entropy values ​​of the enterprise to be analyzed in multiple consecutive years; Construct the industry entropy flow of the industry to which the enterprise to be analyzed belongs; Based on the enterprise entropy flow of the enterprise to be analyzed and the industry entropy flow of the corresponding industry, the internal and external entropy flow analysis results of the enterprise to be analyzed are obtained.

9. The management entropy model optimization method based on AI construction according to claim 2 is characterized in that: After obtaining the management entropy value of the enterprise to be analyzed, it also includes: Determine the relative position of the management entropy value of the enterprise to be analyzed within its industry; Based on the relative position of the management entropy value of the enterprise to be analyzed in its industry and the industry performance standard value, the enterprise score value of the enterprise to be analyzed is obtained.

10. The management entropy model optimization system built based on AI is characterized by: include: The system construction module is used to construct a multi-dimensional evaluation index system for enterprise management entropy evaluation in each industry based on all enterprise evaluation dimensions in each industry and all evaluation indicators under each enterprise evaluation dimension in each industry; The weight optimization module is used to determine the optimized dimension weight of each enterprise evaluation dimension in the multi-dimensional evaluation index system of enterprise management entropy evaluation in each industry and the optimized indicator weight of each evaluation indicator under each enterprise evaluation dimension in each industry based on the AI ​​algorithm; The model optimization module is used to construct a management entropy measurement optimization model for each industry based on the optimized dimension weights of each enterprise evaluation dimension in each industry and the optimized indicator weights of each evaluation indicator under each enterprise evaluation dimension in each industry.

Images