Enterprise development competitiveness evaluation system and method

Abstract

The invention discloses an enterprise development competitiveness evaluation system and method, relates to the technical field of enterprise informatization management, and solves the problems that in the prior art, self-adaptive weight adjustment according to enterprises in different development environments and enterprise types is not considered, so that enterprise development competitiveness analysis is not accurate and comprehensive enough, and the enterprise development competitiveness evaluation efficiency is improved. And the efficiency of the enterprise development competitiveness evaluation system is low. An industry trend level is generated according to an industry category and external data, a basic weight is generated according to the industry trend level, a plurality of enterprise parameter forces are generated according to the basic weight and enterprise parameter data, enterprise development competitiveness is calculated according to the enterprise parameter forces, an enterprise development evaluation report is generated, and the basic weights of each industry are updated within a certain time period. When the actual weight is dynamically adjusted, the change of enterprise characteristics is considered, smoothing processing is carried out according to the current basic weight, severe fluctuation is prevented, the evaluation accuracy and comprehensiveness of enterprise development competitiveness are improved, and then the efficiency of an enterprise development competitiveness evaluation system is improved.

Description

Technical Field

[0001] This application belongs to the technical field of enterprise informatization management, specifically an enterprise development competitiveness evaluation system and method. Background Art

[0002] Enterprise development competitiveness refers to the comprehensive ability of an enterprise in a competitive market to provide products and services to the market more effectively than other enterprises and obtain profits and prestige. This ability is achieved by an enterprise through cultivating its own resources and capabilities, acquiring external accessible resources, and comprehensively utilizing them on the basis of creating value for customers. By clarifying market positioning, strengthening innovation and R & D, optimizing quality management and brand building, strengthening customer relationship management, optimizing internal management, strengthening supply chain management, and actively fulfilling social responsibilities, etc., an enterprise can continuously enhance its core competitiveness and achieve sustainable and stable development.

[0003] The prior art (a patent for invention with the publication number of CN112790559B) discloses an intelligent analysis method and system for enterprise comprehensive competitiveness and development trend, including the following steps: S1, displaying and analyzing the core products of the enterprise; S2, analyzing the number of intellectual properties of the enterprise; S3, analyzing the market share of the enterprise's products; S4, analyzing the brand power of the enterprise; S5, analyzing the sales growth power of the enterprise's products; S6, scoring the analysis results in the above S1 - S5 and obtaining the enterprise comprehensive competitiveness and development trend score according to weighting.

[0004] The above - mentioned case obtains the enterprise development competitiveness by analyzing and weighting various parts of the enterprise's data, lacking consideration of adaptively adjusting weights according to different development environments and enterprise types of the enterprise, and the weights between several data will also change over time, making the analysis of enterprise development competitiveness inaccurate and incomplete, resulting in low efficiency of the enterprise development competitiveness evaluation system; therefore, the enterprise development competitiveness evaluation system still needs further improvement. Summary of the Invention

[0005] This application aims to solve at least one of the technical problems existing in the prior art; for this purpose, this application proposes an enterprise development competitiveness evaluation system and method, which is used to solve the technical problem that the prior art lacks consideration of adaptively adjusting weights according to different development environments and enterprise types of the enterprise, making the analysis of enterprise development competitiveness inaccurate and incomplete, resulting in low efficiency of the enterprise development competitiveness evaluation system.

[0006] To achieve the above - mentioned purpose, the first aspect of this application provides an enterprise development competitiveness evaluation system and method, including: a data acquisition module, a data analysis module, a visualization module, and a database;

[0007] The data acquisition module: obtains enterprise data and external data through data acquisition devices; the enterprise data includes enterprise ID, industry category, enterprise characteristic data, and several enterprise parameter data; the external data includes international data, domestic data, and policy data;

[0008] The data analysis module: generates an industry trend level and trend reasons based on the industry category and external data; generates several enterprise parameter forces based on several enterprise parameter data; calculates the enterprise development competitiveness based on several enterprise parameter forces and then generates an enterprise development evaluation report;

[0009] The visualization module: displays the development evaluation report through the visualization module;

[0010] The database is used to store the data of each module and the historical data required for training the model.

[0011] Through the above steps, this application periodically updates the basic weights of multiple industries within a certain time period. During the dynamic adjustment process of calculating the actual weights, it fully incorporates the change factors caused by the unique nature of the enterprise, and implements a smooth transition strategy based on the current basic weights, aiming to avoid drastic fluctuations. This not only significantly improves the accuracy and comprehensiveness of the evaluation of enterprise development competitiveness, but also further optimizes the overall operation efficiency of the enterprise development competitiveness evaluation system.

[0012] Further, the generation of the industry trend level and trend reasons based on the industry category and external data includes:

[0013] Obtain the recording time, industry type, and external data, as well as the historical industry type, historical external data, and historical recording time;

[0014] Integrate the industry type and external data into several trend prediction sequences in chronological order of the recording time;

[0015] Input several trend prediction sequences into the industry trend level prediction model to obtain the industry trend level and trend reasons; the industry trend level prediction model is constructed through a machine learning model.

[0016] Further, the construction of the industry trend level prediction model through a machine learning model includes:

[0017] Obtain several historical recording times and their corresponding historical industry types, historical external data, historical industry trend levels, and trend reasons;

[0018] Integrate several historical recording times and their corresponding historical industry types, historical external data, historical industry trend levels, and trend reasons into several historical trend prediction sequences and their corresponding historical industry trend levels and trend reasons;

[0019] Divide several historical trend prediction sequences and their corresponding historical industry trend levels and trend reasons into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set;

[0020] Select a machine learning model as the basic model;

[0021] Train the basic model with the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0022] Verify the pre-trained model on the test set, and finally obtain an industry trend level prediction model that inputs several trend prediction sequences and outputs the industry trend level and trend reasons.

[0023] Furthermore, the generation of several enterprise parameter forces according to several enterprise parameter data includes:

[0024] Obtain several enterprise parameter data; the enterprise parameter data includes technical data, product data, market data, resource data, and operation data;

[0025] Obtain the industry category and several weight labels and basic weights within the corresponding current time period; the weight labels and basic weights are generated through historical basic weights and industry trend levels;

[0026] Obtain enterprise characteristic data; the enterprise characteristic data includes scale characteristics, background characteristics, capital structure, ability characteristics, and market characteristics;

[0027] Input the enterprise characteristic data into the enterprise weight correction model to obtain several correction labels and correction coefficients; the correction labels are consistent with the weight labels; the enterprise weight correction model is constructed through an artificial intelligence model;

[0028] Generate several enterprise parameter forces according to several correction labels and their corresponding correction coefficients, basic weights, and enterprise parameter data.

[0029] Furthermore, the generation of the weight labels and basic weights through historical basic weights and industry trend levels includes:

[0030] Obtain several historical weight labels, historical basic weights, and industry trend levels;

[0031] Input several historical weight labels, historical basic weights, and industry trend levels into the basic weight adjustment model to obtain weight labels and basic weights;

[0032] Among them, the basic weight adjustment model is constructed through a machine learning model, including:

[0033] Obtain a number of historical weight tags, historical basic weights, historical industry trend levels, and their corresponding historical weight tags and basic weights;

[0034] Divide a number of historical weight tags, historical basic weights, historical industry trend levels, and their corresponding historical weight tags and basic weights into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set;

[0035] Select a machine learning model as the basic model;

[0036] Train the basic model with the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0037] Verify the pre-trained model on the test set, and finally obtain a basic weight adjustment model that inputs a number of historical weight tags, historical basic weights, and industry trend levels and outputs a number of weight tags and a number of basic weights.

[0038] Furthermore, the enterprise weight correction model is constructed through an artificial intelligence model, including:

[0039] Obtain a number of historical enterprise characteristic data and their corresponding historical correction tags and correction coefficients;

[0040] Divide a number of historical enterprise characteristic data and their corresponding historical correction tags and correction coefficients into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set;

[0041] Select an artificial intelligence model as the basic model;

[0042] Train the basic model with the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0043] Verify the pre-trained model on the test set, and finally obtain an enterprise weight correction model that inputs enterprise characteristic data and outputs correction tags and correction coefficients.

[0044] Furthermore, generating a number of enterprise parameter forces according to a number of correction tags and their corresponding correction coefficients, basic weights, and enterprise parameter data includes:

[0045] Obtain the enterprise ID and its corresponding number of correction tags, correction coefficient XX t,j , basic weight JQ t , enterprise parameter data, and the actual weight SQ in the previous time period t-1 ;

[0046] Through the formula Calculate the weight correction amount ΔXL corresponding to several correction labels t,i,j ; where t represents the time period number, i represents the number corresponding to the correction label, and j represents the number corresponding to each parameter within the correction label; α is the enterprise characteristic influence intensity, α ∈ (0, 1); ω is the state damping coefficient, ω ∈ (0, 1); β is the mean reversion intensity, β ∈ (0, 1);

[0047] Through the formula SQ t,i,j = JQ t,i,j + ΔXL t,i,j Calculate the actual weight SQ t,i,j ;

[0048] Judge whether the actual weight exceeds its corresponding weight range;

[0049] Yes, when the actual weight is greater than the maximum value in its corresponding weight range, make the actual weight equal to the maximum value in the weight range; when the actual weight is less than the minimum value in its corresponding weight range, make the actual weight equal to the minimum value in the weight range;

[0050] No, judge whether the sum of several actual weights within the same correction label is equal to 1; Yes, do nothing; No, obtain several actual weights through normalization;

[0051] Obtain the full score of the enterprise development competitiveness standard and the comprehensive label in the correction label;

[0052] Through the formula QLM t,j = QJM × SQ t,j Calculate the full score QLM of several enterprise parameter forces t,j ;

[0053] Calculate several enterprise parameter forces according to the full score of the enterprise parameter force standard and the enterprise parameter data.

[0054] This application flexibly adjusts the actual weights corresponding to various types of enterprises based on enterprise characteristics and basic weights. In the process of calculating the actual weights, in order to ensure that the corrected weights neither exceed their preset ranges nor violate the sum of all weights within the same type being equal to 1, weight range limitation and normalization processing measures are taken, improving the accuracy of the actual weights, and thus enhancing the accuracy and comprehensiveness of the enterprise development competitiveness evaluation.

[0055] Furthermore, the calculating several enterprise parameter forces according to the full score of the enterprise parameter force standard and the enterprise parameter data includes:

[0056] Obtain several enterprise parameter force standards full marks QLM, several enterprise parameter data and their corresponding actual weights; the enterprise parameter data includes technology data, product data, market data, resource data and operation data; the technology data includes the proportion of R & D expenses, the conversion rate of technological achievements and the growth rate of emerging technology patent applications; the product data includes existing product types, future product types, main product types and product iteration cycles; the market data includes regional market share, market price, number of customers, number of large customers, customer lifetime value ratio and brand search heat value; the resource data includes the supply chain resource sharing rate, number of strategic alliances and asset securitization ratio; the operation data includes the growth rate of per capita revenue, operation expense rate and overdue rate of accounts receivable;

[0057] Perform parameter standardization on several parameters in the technology data, product data, market data, resource data and operation data to obtain several standard parameters in the technology data, product data, market data, resource data and operation data;

[0058] Through the formula JYL t,i =QLM t,i ×(Σ j BC t,i,j ×SQ t,i,j ) Calculate several enterprise parameter forces JYL t,i ; where BC t,i,j represents the standard parameter value of the jth parameter under the ith weight label in the tth time period after parameter standardization; the enterprise parameter forces include technology extension force, product competitiveness, market penetration force, resource integration force and comprehensive operation force.

[0059] Further, after calculating the enterprise development competitiveness according to several enterprise parameter forces, generate an enterprise development evaluation report, including:

[0060] Obtain several enterprise parameter forces, enterprise data, industry trend levels and trend reasons;

[0061] Calculate the enterprise development competitiveness by summing several enterprise parameter forces;

[0062] Integrate the enterprise ID and its corresponding enterprise data, industry trend level, trend reason, enterprise parameter force and enterprise development competitiveness into comprehensive analysis data;

[0063] Input the comprehensive analysis data into the report evaluation large model to obtain the enterprise ID and its corresponding enterprise development evaluation report; the report evaluation large model is constructed through a pre-trained large model.

[0064] Further, the report evaluation large model is constructed through a pre-trained large model, including:

[0065] Obtain data in the field of enterprise evaluation and a pre-trained large model; the data in the field of enterprise evaluation includes enterprise data, industry trend levels, reasons for trends, enterprise parameter forces, and enterprise development competitiveness and their corresponding enterprise development evaluation reports;

[0066] Integrate enterprise data, industry trend levels, reasons for trends, enterprise parameter forces, and enterprise development competitiveness and their corresponding enterprise development evaluation reports into a number of historical comprehensive analysis data and a number of historical enterprise development evaluation reports;

[0067] Divide a number of historical comprehensive analysis data and a number of historical enterprise development evaluation reports into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set;

[0068] Select a pre-trained large model as the basic model;

[0069] Train the basic model with the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0070] Verify the pre-trained model on the test set, and finally obtain a report evaluation large model that takes comprehensive analysis data as input and outputs enterprise IDs and enterprise development evaluation reports.

[0071] Another aspect of the present invention provides a method for evaluating enterprise development competitiveness, including:

[0072] S0: Obtain enterprise data and external data; the enterprise data includes enterprise IDs, industry categories, enterprise characteristic data, and a number of enterprise parameter data;

[0073] S1: Generate industry trend levels and reasons for trends based on industry categories and external data;

[0074] S2: Generate a number of enterprise parameter forces based on a number of enterprise parameter data;

[0075] S3: Calculate enterprise development competitiveness based on a number of enterprise parameter forces and then generate an enterprise development evaluation report;

[0076] S4: Display the development evaluation report through a visualization module.

[0077] Compared with the prior art, the beneficial effects of the present application are:

[0078] 1. This application generates industry trend levels and trend reasons based on industry categories and external data; generates a number of enterprise parameter forces based on a number of enterprise parameter data; calculates the enterprise development competitiveness based on a number of enterprise parameter forces, and then generates an enterprise development evaluation report. It updates a number of basic weights for each industry within a certain time period, and when making the final actual weight dynamic adjustment, it considers the changes brought by enterprise characteristics, and at the same time performs smoothing processing according to the current basic weights to prevent violent fluctuations, improve the evaluation accuracy and comprehensiveness of enterprise development competitiveness, and thus improve the efficiency of the enterprise development competitiveness evaluation system.

[0079] 2. This application analyzes the industry types, domestic and foreign relevant data, and policy data to judge the future development trends of the current industries. For industries with different development trends, the methods and corresponding weight coefficients adopted in calculating enterprise development competitiveness are different, so as to better analyze various data affecting enterprise development competitiveness and improve the comprehensive evaluation effect of enterprise development competitiveness.

[0080] 3. This application generates the basic weights within the current time period based on historical basic weights and industry trend levels, and updates the basic weights within a certain time range to prevent the situation where the basic weights still maintain the original data in the case of policy reforms and social environment changes, resulting in the basic weights not being able to keep up with the times and lacking adaptability and accuracy.

[0081] 4. This application adaptively adjusts the actual weights corresponding to enterprise types according to enterprise characteristics and basic weights. At the same time, on the basis of calculating the actual weights, in order to prevent the corrected actual weights from exceeding their corresponding weight ranges and the sum of several weights of the same type after correction not being 1, through restricting the weight range and normalization operations, the actual weights are made more accurate, and the evaluation accuracy and comprehensiveness of enterprise development competitiveness are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0083] Figure 1 It is a schematic diagram of the principle of the enterprise development competitiveness evaluation system of the present application;

[0084] Figure 2 It is a flowchart for generating the enterprise development evaluation report of the present application;

[0085] Figure 3This is the flowchart of the enterprise development competitiveness evaluation method of the present application. Detailed implementation manners

[0086] Next, the technical solutions of the present application will be clearly and completely described in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0087] Please refer to Figure 1 , the first aspect embodiment of the present application provides an enterprise development competitiveness evaluation system, including: a data collection module, a data analysis module, a visualization module, and a database;

[0088] Data collection module: Obtain enterprise data and external data through data collection devices; enterprise data includes enterprise ID, industry category, enterprise characteristic data, and several enterprise parameter data; external data includes international data, domestic data, and policy data; data collection devices include several sensors, etc.

[0089] Data analysis module: Generate an industry trend level and a trend reason according to the industry category and external data; the industry trend level refers to the future development trend of the current industry, and the trend reason refers to the reason for the future development trend; generate several enterprise parameter forces according to several enterprise parameter data, and the enterprise parameter force refers to the index data required for calculating the enterprise development competitiveness; generate an enterprise development evaluation report after calculating the enterprise development competitiveness according to several enterprise parameter forces, and the enterprise development evaluation report refers to an analysis report of all data on the future development competitiveness of the current enterprise.

[0090] Visualization module: Display the development evaluation report through the visualization module;

[0091] The database is used to store the data of each module and the historical data required for training the model.

[0092] Generating an industry trend level and a trend reason according to the industry category and external data in this embodiment includes:

[0093] Obtain the recording time, industry type, and external data, as well as the historical industry type, historical external data, and historical recording time;

[0094] Integrate the industry type and external data into several trend prediction sequences in chronological order of the recording time;

[0095] Input several trend prediction sequences into the industry trend level prediction model to obtain the industry trend level and the trend reason; the industry trend level prediction model is constructed through a machine learning model.

[0096] The industry trend level prediction model in this embodiment is constructed through a machine learning model, including:

[0097] Obtain a number of historical record times and their corresponding historical industry types, historical external data, historical industry trend levels, and trend reasons;

[0098] Integrate a number of historical record times and their corresponding historical industry types, historical external data, historical industry trend levels, and trend reasons into a number of historical trend prediction sequences and their corresponding historical industry trend levels and trend reasons;

[0099] Divide a number of historical trend prediction sequences and their corresponding historical industry trend levels and trend reasons into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; the ratio among the training set, the test set, and the validation set is 7:2:1;

[0100] Select a machine learning model as the basic model; the machine learning model includes an LSTM model, etc.;

[0101] Train the basic model through the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0102] Verify the pre-trained model on the test set, and finally obtain an industry trend level prediction model that inputs a number of trend prediction sequences and outputs an industry trend level and trend reasons.

[0103] This embodiment deeply analyzes the industry type, domestic and foreign relevant data, and policy orientation to accurately predict the future development trends of each industry; for industries with different trends, differential methods and corresponding weight coefficients are adopted to evaluate the enterprise development competitiveness, which can more comprehensively and meticulously examine the multiple data affecting the enterprise development competitiveness, thereby effectively improving the accuracy and depth of the comprehensive evaluation.

[0104] Generating a number of enterprise parameter forces according to a number of enterprise parameter data in this embodiment includes:

[0105] Obtain a number of enterprise parameter data; the enterprise parameter data includes technology data, product data, market data, resource data, and operation data;

[0106] Obtain the industry category and a number of weight labels and basic weights within the corresponding current time period; the weight labels and basic weights are generated through historical basic weights and industry trend levels;

[0107] Obtain enterprise characteristic data; the enterprise characteristic data includes scale characteristics, background characteristics, capital structure, ability characteristics, and market characteristics;

[0108] Input enterprise characteristic data into the enterprise weight correction model to obtain several correction labels and correction coefficients; the correction labels are consistent with the weight labels; the enterprise weight correction model is constructed through an artificial intelligence model; the correction labels are expressed as labels for the weight coefficients of the weight labels.

[0109] Generate several enterprise parameter forces based on several correction labels and their corresponding correction coefficients, basic weights, and enterprise parameter data.

[0110] In this embodiment, the weight labels and basic weights are generated through historical basic weights and industry trend levels, including:

[0111] Obtain several historical weight labels, historical basic weights, and industry trend levels.

[0112] Input several historical weight labels, historical basic weights, and industry trend levels into the basic weight adjustment model to obtain weight labels and basic weights; the basic weights corresponding to each industry in different periods will be different, and the basic weights between different industries will also be different.

[0113] Among them, the basic weight adjustment model is constructed through a machine learning model, including:

[0114] Obtain several historical weight labels, historical basic weights, historical industry trend levels, and their corresponding historical weight labels and basic weights; since calculating the enterprise development competitiveness requires many parameters, which belong to calculating different indicators, the weights between multiple parameters under each indicator will also be different.

[0115] Divide several historical weight labels, historical basic weights, historical industry trend levels, and their corresponding historical weight labels and basic weights into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; the ratio among the training set, the test set, and the validation set is 7:2:1.

[0116] Select a machine learning model as the basic model; the machine learning model includes a neural network model, etc.

[0117] Train the basic model through the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model.

[0118] Verify the pre-trained model on the test set, and finally obtain a basic weight adjustment model that inputs several historical weight labels, historical basic weights, and industry trend levels and outputs several weight labels and several basic weights.

[0119] In this embodiment, the basic weight for the current time period is carefully formulated based on historical basic weights and industry trend levels, and is updated within a specific time period to prevent the basic weight from maintaining the original data during policy adjustments or social environment changes. Thus, it ensures that the basic weight can keep pace with the times, has the ability of self - adjustment and high accuracy, effectively prevents the basic weight from getting out of touch with the times, and guarantees its adaptability and precision when evaluating the development competitiveness of enterprises.

[0120] The enterprise weight correction model in this embodiment is constructed through an artificial intelligence model, including:

[0121] Obtain a number of historical enterprise characteristic data and their corresponding historical correction labels and correction coefficients;

[0122] Divide a number of historical enterprise characteristic data and their corresponding historical correction labels and correction coefficients into training data, validation data, and test data; perform data pre - processing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; the ratio among the training set, the test set, and the validation set is 7:2:1;

[0123] Select an artificial intelligence model as the basic model; the artificial intelligence model includes BP models, etc.;

[0124] Train the basic model through the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre - trained model;

[0125] Verify the pre - trained model on the test set, and finally obtain an enterprise weight correction model that takes enterprise characteristic data as input and outputs correction labels and correction coefficients.

[0126] Generating a number of enterprise parameter forces according to a number of correction labels and their corresponding correction coefficients, basic weights, and enterprise parameter data in this embodiment includes:

[0127] Obtain the enterprise ID and its corresponding number of correction labels, correction coefficient XX t,j 、basic weight JQ t 、enterprise parameter data, and the actual weight SQ in the previous time period t-1 ;

[0128] Through the formula Calculate the weight correction amount ΔXL corresponding to a number of correction labels t,i,j, the weight correction amount can be positive or negative. When the weight correction amount is negative, it indicates that the value of the basic weight needs to be decreased; when the weight correction amount is positive, it indicates that the value of the basic weight needs to be increased. Among them, t represents the time period number, i represents the number corresponding to the correction label, and j represents the number corresponding to each parameter within the correction label; α is the influence intensity of enterprise characteristics, α ∈ (0, 1), and the specific value is set according to experience. In this embodiment, α is set to 0.7; ω is the state damping coefficient, and the setting of ω is to control the correction attenuation speed when the current weight deviates from the benchmark, ω ∈ (0, 1), and the specific value is set according to experience. In this embodiment, ω is set to 0.5; β is the mean reversion intensity, β ∈ (0, 1), and the specific value is set according to experience. In this embodiment, β is set to 0.3, and the setting of β is to adjust the speed of pulling the weight back to the industry benchmark to prevent long-term deviation; the correction direction is determined by enterprise characteristics, and the current weight state controls the correction intensity, so that the actual weight can change adaptively; when the basic weight and the actual weight in the previous time period are fixed, as the correction coefficient increases, the weight correction amount will also increase;

[0129] Through the formula SQ t,i,j = JQ t,i,j +ΔXL t,i,j Calculate the actual weight SQt ,i,j ; Add it to the weight correction amount on the basis of the basic weight to obtain the actual weight;

[0130] Judge whether the actual weight exceeds its corresponding weight range; This step is to prevent the situation that the weight correction degree is too large during the weight adjustment process, resulting in non-compliance with the actual problem.

[0131] Yes, when the actual weight is greater than the maximum value in its corresponding weight range, let the actual weight be equal to the maximum value in the weight range; when the actual weight is less than the minimum value in its corresponding weight range, let the actual weight be equal to the minimum value in the weight range;

[0132] No, judge whether the sum of several actual weights within the same correction label is equal to 1; Yes, do nothing; No, obtain several actual weights through normalization; Among them, the normalization method can be calculated as follows, and the normalized actual weight is

[0133] Obtain the full score QJM of the enterprise development competitiveness standard and the comprehensive label in the correction label. The value of QJM can be set to 10 points or 100 points, and the specific value can be set according to experience. In this embodiment, QJM is set to 100 points; The comprehensive label refers to the label corresponding to the actual weight among several enterprise parameter forces;

[0134] Through the formula QLM t,j=QJM×SQ t,j Calculate the full score QLM of the enterprise parameter force standard for several enterprises t,j ; As the actual weight increases, the corresponding full score of the enterprise parameter force standard will gradually increase;

[0135] Calculate the enterprise parameter forces of several enterprises based on the full score of the enterprise parameter force standard and the enterprise parameter data

[0136] In this embodiment, calculating the enterprise parameter forces of several enterprises based on the full score of the enterprise parameter force standard and the enterprise parameter data includes:

[0137] Obtain the full score QLM of the enterprise parameter force standard for several enterprises, the enterprise parameter data and their corresponding actual weights; the enterprise parameter data includes technical data, product data, market data, resource data and operation data; the technical data includes the proportion of R & D expenses, the conversion rate of technical achievements and the growth rate of emerging technology patent applications, etc.; the product data includes the existing product types, future product types, main product types and product iteration cycles, etc.; the market data includes the regional market share, market price, number of customers, number of large customers, customer lifetime value ratio and brand search heat value, etc.; the resource data includes the supply chain resource sharing rate, number of strategic alliances and asset securitization ratio, etc.; the operation data includes the growth rate of per capita revenue, operation expense rate and overdue rate of accounts receivable, etc.;

[0138] Perform parameter standardization on several parameters in the technical data, product data, market data, resource data and operation data to obtain several standard parameters in the technical data, product data, market data, resource data and operation data; for example, for the standardization of the proportion of R & D expenses in the technical data: obtain it through min(actual proportion / industry benchmark value, 1); for the standardization of the technology conversion rate: obtain it through the number of successfully converted projects / total number of R & D projects; for the standardization of the patent growth rate: obtain it through max((number of patents in this time period - number of patents in the previous time period) / number of patents in the previous time period, 0); the time period is set according to experience, and can be set to 1 quarter or 1 year; data such as the industry benchmark value and the number of patents in the previous time period can be obtained directly;

[0139] Through the formula JYL t,i =QLM t,i ×(∑ j BC t,i,j ×SQ t,i,j ) Calculate the enterprise parameter forces JYL of several enterprises t,i ; Among them, BC t,i,jIt represents the standard parameter value of the j-th parameter after parameter standardization under the i-th weight label in the t-th time period; when the standard full score of each enterprise's parameter force is fixed, the enterprise's parameter force increases with the increase of its corresponding standard parameter value and actual weight; the enterprise's parameter force includes technology extension force, product competitiveness, market penetration force, resource integration force, and comprehensive operation force.

[0140] In this embodiment, by performing standardization operations on several parameters and generating the actual full score values of each part of the enterprise development competitiveness in advance, based on the standardized several parameters and their corresponding actual weights on the basis of each actual full score value, the calculation of the enterprise development competitiveness index is faster and more accurate.

[0141] Please refer to Figure 2 , after calculating the enterprise development competitiveness according to several enterprise parameter forces in this embodiment, an enterprise development evaluation report is generated, including:

[0142] Obtain several enterprise parameter forces, enterprise data, industry trend levels, and trend reasons;

[0143] Calculate the enterprise development competitiveness by summing several enterprise parameter forces; since the full score values of each part have been assigned in advance when calculating several enterprise parameter forces, after calculating the scores that each department can obtain under the full score value, directly sum them to obtain the enterprise development competitiveness;

[0144] Integrate the enterprise ID and its corresponding enterprise data, industry trend level, trend reason, enterprise parameter force, and enterprise development competitiveness into comprehensive analysis data;

[0145] Input the comprehensive analysis data into the report evaluation large model to obtain the enterprise ID and its corresponding enterprise development evaluation report; the report evaluation large model is constructed through a pre-trained large model.

[0146] The report evaluation large model in this embodiment is constructed through a pre-trained large model, including:

[0147] Obtain enterprise evaluation domain data and a pre-trained large model; the enterprise evaluation domain data includes enterprise data, industry trend levels, trend reasons, enterprise parameter forces, and enterprise development competitiveness and their corresponding enterprise development evaluation reports;

[0148] Integrate enterprise data, industry trend levels, trend reasons, enterprise parameter forces, and enterprise development competitiveness and their corresponding enterprise development evaluation reports into several historical comprehensive analysis data and several historical enterprise development evaluation reports;

[0149] Divide a number of historical comprehensive analysis data and a number of historical enterprise development evaluation reports into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; the ratio among the training set, the test set, and the validation set is 7:2:1;

[0150] Select a pre-trained large model as the basic model; the pre-trained large language models include chatGPT, etc.;

[0151] Train the basic model with the training set, and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model;

[0152] Verify the pre-trained model on the test set, and finally obtain a report evaluation large model that takes the comprehensive analysis data as input and outputs the enterprise ID and the enterprise development evaluation report.

[0153] Please refer to Figure 3 , another embodiment of the present application provides an enterprise development competitiveness evaluation method, including:

[0154] S0: Obtain enterprise data and external data; the enterprise data includes enterprise ID, industry category, enterprise characteristic data, and a number of enterprise parameter data;

[0155] S1: Generate an industry trend level and a trend reason according to the industry category and external data;

[0156] S2: Generate a number of enterprise parameter forces according to a number of enterprise parameter data;

[0157] S3: Calculate the enterprise development competitiveness according to a number of enterprise parameter forces and then generate an enterprise development evaluation report;

[0158] S4: Display the development evaluation report through a visualization module.

[0159] Some of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is obtained by software simulation of a large amount of collected data to get a formula closest to the real situation; the preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0160] Working principle of this application: By obtaining enterprise data and external data; the enterprise data includes enterprise ID, industry category, enterprise characteristic data and several enterprise parameter data; generating industry trend level and trend reasons according to the industry category and external data; generating several enterprise parameter forces according to several enterprise parameter data; calculating enterprise development competitiveness according to several enterprise parameter forces and then generating an enterprise development evaluation report; displaying the development evaluation report through a visualization module, updating several basic weights of each industry within a certain time period, and considering the changes brought by enterprise characteristics when making the final actual weight dynamic adjustment, and at the same time performing smoothing processing according to the current basic weights to prevent violent fluctuations, improving the evaluation accuracy and comprehensiveness of enterprise development competitiveness, thereby enhancing the efficiency of the enterprise development competitiveness evaluation system, and avoiding the problem that the prior art lacks consideration of adaptively adjusting weights according to the enterprise in different development environments and enterprise types, resulting in inaccurate and incomplete analysis of enterprise development competitiveness and low efficiency of the enterprise development competitiveness evaluation system.

[0161] The above embodiments are only used to illustrate the technical method of this application and not to limit it. Although this application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of this application can be modified or equivalently replaced without departing from the spirit and scope of the technical method of this application.

Claims

1. Enterprise development competitiveness evaluation system, characterized in that, Including: A data acquisition module, a data analysis module, a visualization module, and a database; The data acquisition module: obtains enterprise data and external data through data acquisition devices; the enterprise data includes enterprise ID, industry category, enterprise characteristic data, and a number of enterprise parameter data; the external data includes international data, domestic data, and policy data; The data analysis module: generates an industry trend level and a trend reason based on the industry category and external data; Generates a weight label and a basic weight through a machine learning model according to the industry trend level, and generates a number of enterprise parameter forces according to the basic weight and a number of enterprise parameter data; Calculates the enterprise development competitiveness based on a number of enterprise parameter forces and then generates an enterprise development evaluation report.

2. The enterprise development competitiveness evaluation system according to claim 1, wherein The generating of the industry trend level and the trend reason based on the industry category and external data includes: Obtains the recording time, industry type, and external data, as well as the historical industry type, historical external data, and historical recording time; Integrates the industry type and external data into a number of trend prediction sequences in chronological order of the recording time; Inputs the number of trend prediction sequences into an industry trend level prediction model to obtain the industry trend level and the trend reason; the industry trend level prediction model is constructed through a machine learning model.

3. The enterprise development competitiveness evaluation system according to claim 2, characterized in that, The construction of the industry trend level prediction model through a machine learning model includes: Obtains a number of historical recording times and their corresponding historical industry types, historical external data, and historical industry trend levels and trend reasons; Integrates a number of historical recording times and their corresponding historical industry types, historical external data, and historical industry trend levels and trend reasons into a number of historical trend prediction sequences and their corresponding historical industry trend levels and trend reasons; Divides a number of historical trend prediction sequences and their corresponding historical industry trend levels and trend reasons into training data, validation data, and test data; performs data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; Selects a machine learning model as the basic model; Trains the basic model through the training set and adjusts the learning rate and hyperparameters on the validation set to obtain a pre-trained model; Verifies the pre-trained model on the test set, and finally obtains an industry trend level prediction model that inputs a number of trend prediction sequences and outputs the industry trend level and the trend reason.

4. The enterprise development competitiveness evaluation system according to claim 1, wherein The generating of the weight label and the basic weight according to the industry trend level includes: Obtains a number of historical weight labels, historical basic weights, and industry trend levels; Inputs a number of historical weight labels, historical basic weights, and industry trend levels into a basic weight adjustment model to obtain the weight label and the basic weight; Among them, the basic weight adjustment model is constructed through a machine learning model, including: Obtains a number of historical weight labels, historical basic weights, and historical industry trend levels and their corresponding historical weight labels and basic weights; Divides a number of historical weight labels, historical basic weights, and historical industry trend levels and their corresponding historical weight labels and basic weights into training data, validation data, and test data; performs data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; Select a machine learning model as the base model; Train the base model with a training set and adjust the learning rate and hyperparameters on a validation set to obtain a pre-trained model; Validate the pre-trained model on a test set, and finally obtain a base weight adjustment model that takes in a number of historical weight labels, historical base weights, and industry trend levels and outputs a number of weight labels and a number of base weights.

5. The enterprise development competitiveness evaluation system according to claim 4, wherein, The generation of a number of enterprise parameter forces based on the base weights and a number of enterprise parameter data includes: Obtain a number of enterprise parameter data; the enterprise parameter data includes technology data, product data, market data, resource data, and operation data; Obtain the industry category and a number of weight labels and base weights within the corresponding current time period; Obtain enterprise characteristic data; the enterprise characteristic data includes scale characteristics, background characteristics, capital structure, ability characteristics, and market characteristics; Input the enterprise characteristic data into an enterprise weight correction model to obtain a number of correction labels and correction coefficients; the correction labels are consistent with the weight labels; the enterprise weight correction model is constructed through an artificial intelligence model; Generate a number of enterprise parameter forces based on a number of correction labels, their corresponding correction coefficients, base weights, and enterprise parameter data.

6. The enterprise development competitiveness evaluation system according to claim 5, wherein The construction of the enterprise weight correction model through an artificial intelligence model includes: Obtain a number of historical enterprise characteristic data and their corresponding historical correction labels and correction coefficients; Divide a number of historical enterprise characteristic data and their corresponding historical correction labels and correction coefficients into training data, validation data, and test data; perform data preprocessing on the training data, validation data, and test data to obtain a training set, a validation set, and a test set; Select an artificial intelligence model as the base model; Train the base model with the training set and adjust the learning rate and hyperparameters on the validation set to obtain a pre-trained model; Validate the pre-trained model on the test set, and finally obtain an enterprise weight correction model that takes in enterprise characteristic data and outputs correction labels and correction coefficients.

7. The enterprise development competitiveness evaluation system according to claim 5, characterized in that The generation of a number of enterprise parameter forces based on a number of correction labels, their corresponding correction coefficients, base weights, and enterprise parameter data includes: Obtain the enterprise ID and its corresponding number of corrected tags, correction coefficient XX t,j , basic weight JQ t , enterprise parameter data, and the actual weight SQ in the previous time period t-1 ; Through the formula Calculate the weight correction amount ΔXL corresponding to several correction labels t,i,j ; where t represents the time period number, i represents the number corresponding to the correction label, and j represents the number corresponding to each parameter within the correction label; α is the enterprise characteristic influence intensity, α ∈ (0, 1); ω is the state damping coefficient, ω ∈ (0, 1); β is the mean reversion intensity, β ∈ (0, 1); Calculate the actual weight SQ t,i,j = JQ t,i,j + ΔXL t,i,j by the formula SQ t,i,j ; Judge whether the actual weight exceeds its corresponding weight range; If yes, when the actual weight is greater than the maximum value in its corresponding weight range, set the actual weight equal to the maximum value in the weight range; when the actual weight is less than the minimum value in its corresponding weight range, set the actual weight equal to the minimum value in the weight range; If no, judge whether the sum of a number of actual weights within the same correction label is equal to 1; if yes, do nothing; if no, obtain a number of actual weights through normalization; Obtain the full score of the enterprise development competitiveness standard and the comprehensive label in the correction label; Through the formula QLM t,j = QJM × SQ t,j Calculate the full score QLM of the standard of several enterprise parameter forces t,j ; Calculate a number of enterprise parameter forces based on the full score of the enterprise parameter force standard and the enterprise parameter data.

8. The enterprise development competitiveness evaluation system according to claim 7, characterized in that The calculation of a number of enterprise parameter forces based on the full score of the enterprise parameter force standard and the enterprise parameter data includes: Obtain the full score QLM of several enterprise parameter forces, several enterprise parameter data and their corresponding actual weights; the enterprise parameter data includes technology data, product data, market data, resource data and operation data; the technology data includes the proportion of R & D expenses, the conversion rate of technology achievements and the growth rate of emerging technology patent applications; the product data includes existing product types, future product types, main product types and product iteration cycles; the market data includes regional market share, market price, number of customers, number of large customers, customer lifetime value ratio and brand search heat value; the resource data includes the supply chain resource sharing rate, number of strategic alliances and asset securitization ratio; the operation data includes the growth rate of per capita revenue, operation expense rate and overdue rate of accounts receivable. Perform parameter standardization on several parameters in the technology data, product data, market data, resource data and operation data to obtain several standard parameters in the technology data, product data, market data, resource data and operation data. Calculate the enterprise parameter force JYL through the formula t,o =QLM t,i ×(∑ j BC t,i,j ×SQ t,i,j ) t,i ; where BC t,i,j represents the standardized parameter value of the jth parameter under the ith weight label in the tth time period after parameter standardization; the enterprise parameter force includes technology extension force, product competitiveness, market penetration force, resource integration force and comprehensive operation force.

9. The enterprise development competitiveness evaluation system according to claim 1, characterized in that After calculating the enterprise development competitiveness according to several enterprise parameter forces, generate an enterprise development evaluation report, including: Obtain several enterprise parameter forces, enterprise data, industry trend level and trend reasons. Calculate the enterprise development competitiveness by summing several enterprise parameter forces. Integrate the enterprise ID and its corresponding enterprise data, industry trend level, trend reasons, enterprise parameter forces and enterprise development competitiveness into comprehensive analysis data. Input the comprehensive analysis data into the report evaluation large model to obtain the enterprise ID and its corresponding enterprise development evaluation report; the report evaluation large model is constructed by a pre-trained large model.

10. A method for evaluating the competitiveness of enterprise development, which is applied to the enterprise development competitiveness evaluation system according to any one of claims 1-9, and is characterized in that, Including: S0: Obtain enterprise data and external data; the enterprise data includes enterprise ID, industry category, enterprise characteristic data and several enterprise parameter data. S1: Generate the industry trend level and trend reasons according to the industry category and external data. S2: Generate weight labels and basic weights according to the industry trend level; generate several enterprise parameter forces according to the basic weights and several enterprise parameter data. S3: Calculate the enterprise development competitiveness according to several enterprise parameter forces and then generate an enterprise development evaluation report.

Images