Enterprise growth scale prediction method and system
By screening and synthesizing enterprise growth indicators and combining with the reinforcement learning framework, the accuracy of enterprise growth scale prediction is improved and investment decisions are optimized.
Patent Information
- Application Number
- CN202510366994.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-26
- Publication Date
- 2025-07-18
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the existing technology, the forecast of enterprise growth scale is not accurate enough, which affects investment decision-making and risk management.
By screening enterprise growth indicators, calculating entropy value difference coefficients, determining weights, synthesize enterprise growth scale indicators, and inputting them into the trained enterprise growth scale prediction model, combining the reinforcement learning framework to adjust the investment strategy multiple times, and finally obtain the best investment strategy.
Improve the accuracy of corporate growth scale forecasting and optimize investment decisions.
Smart Images
Figure CN120338960A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the technical field of data prediction, and in particular, to a method and system for predicting the growth scale of an enterprise. Background Art
[0002] The significance of the growth scale of an enterprise lies in enhancing market competitiveness, strengthening resource acquisition capabilities, diversifying risks, and boosting brand influence. These advantages help the enterprise to solidify its position in the fierce market competition and achieve sustainable development.
[0003] With the development of the digital economy, investors increasingly attach importance to predicting the growth scale of an enterprise, as it helps them grasp market trends, evaluate investment value, manage investment portfolios, control risks, and identify enterprises with growth potential. However, the current means of predicting the growth scale of an enterprise are not accurate enough. Therefore, how to improve the accuracy of predicting the growth scale of an enterprise has become an urgent problem to be solved. Summary of the Invention
[0004] The embodiments of the present application provide a method and means for predicting the growth scale of an enterprise to improve the accuracy of predicting the growth scale of an enterprise.
[0005] In a first aspect, the embodiments of the present application provide a method for predicting the growth scale of an enterprise, the method comprising: Screening enterprise growth indicators and synthesizing the enterprise growth indicators into an enterprise growth scale indicator; Collecting data according to the enterprise growth scale indicator to generate enterprise growth scale data; Inputting the enterprise growth data into a trained enterprise growth scale prediction model for prediction to obtain an enterprise growth scale prediction result.
[0006] In some possible implementation manners, the screening of the enterprise growth indicators and the synthesizing of the enterprise growth indicators into an enterprise growth scale indicator include: Calculating the difference coefficient of the entropy values of each enterprise growth indicator; Calculating the weight of each enterprise growth indicator through the difference coefficient; Screening the top three enterprise growth indicators with weights and obtaining the enterprise growth scale indicator through weighted summation.
[0007] In some possible implementation manners, the method comprises: Integrating the enterprise growth scale prediction model into a reinforcement learning framework; Obtaining corresponding enterprise growth scale prediction results by adjusting the investment strategy multiple times in the reinforcement learning framework; Comparing the enterprise growth scale prediction results to obtain the optimal investment strategy.
[0008] In some possible embodiments, it is characterized in that: The state of the reinforcement learning framework consists of a set of key feature variables; among them, the set of key variable features includes market growth rate, enterprise profitability, economic indicators, and the prediction result of the historical enterprise growth scale; The adjusted investment strategy is a set of investment actions; among them, the investment actions include the percentages of increasing, decreasing, or maintaining investment in different assets.
[0009] In some possible embodiments, it is characterized in that the method further includes: Preprocess the attribute data of the enterprise; including data cleaning, screening key feature variables, and data integration.
[0010] In a second aspect, an embodiment of the present application provides an enterprise growth scale prediction system, and the system includes: An index screening unit, configured to screen enterprise growth indexes and synthesize the enterprise growth indexes into an enterprise growth scale index; A data acquisition unit, configured to acquire data according to the enterprise growth scale index and generate enterprise growth scale data; A growth prediction unit, configured to input the enterprise growth data into a trained enterprise growth scale prediction model for prediction to obtain an enterprise growth scale prediction result.
[0011] In some possible embodiments, The index screening unit is configured to calculate the difference coefficient of the entropy values of each enterprise growth index; The index screening unit is further configured to calculate the weights of each enterprise growth index through the difference coefficient; The index screening unit is further configured to screen the top three enterprise growth indexes with weights and obtain the enterprise growth scale index through weighted summation.
[0012] In some possible embodiments, the system further includes a reinforcement learning unit; The reinforcement learning unit is configured to integrate the enterprise growth scale prediction model into a reinforcement learning framework; The reinforcement learning unit is further configured to obtain a corresponding enterprise growth scale prediction result by adjusting the investment strategy multiple times in the reinforcement learning framework; The reinforcement learning unit is further configured to compare the enterprise growth scale prediction results to obtain the best investment strategy.
[0013] In some possible embodiments, The state of the reinforcement learning framework consists of a set of key feature variables; among them, the set of key variable features includes market growth rate, enterprise profitability, economic indicators, and the prediction result of the historical enterprise growth scale; The adjusted investment strategy is a set of investment actions; among them, the investment actions include multiple percentages of increasing, decreasing, or maintaining investments in different assets.
[0014] In some possible implementation manners, the system further includes a data processing unit; The data processing unit is used to preprocess the attribute data of the enterprise; including data cleaning, screening key feature variables, and data integration.
[0015] The beneficial effects of this application are: In the embodiment of this application, by screening the enterprise growth indicators, the enterprise growth indicators are synthesized into an enterprise growth scale indicator, then data collection is performed according to the enterprise growth scale indicator to generate enterprise growth scale data, and finally the enterprise growth data is input into a trained enterprise growth scale prediction model for prediction to obtain an enterprise growth scale prediction result, thereby achieving the improvement of the accuracy of enterprise growth scale prediction. Description of the Drawings
[0016] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.
[0017] Figure 1 It is a flowchart of the enterprise growth scale prediction method provided by the embodiment of this application; Figure 2 It is a schematic diagram of the enterprise growth scale prediction system provided by the embodiment of this application. Detailed Embodiments
[0018] To make the purpose, technical solutions, and advantages of this application clearer, the following will refer to the drawings in the embodiments of this application and describe the technical solutions of this application in detail through implementation manners. Obviously, the described embodiments are some embodiments of this application, rather than all embodiments. Without conflict, the embodiments and the features in the embodiments of this application can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.
[0019] It should be noted that in the accompanying drawings, the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout; in the description of the present application, the terms "center", "longitudinal", "transverse", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the protection scope of the present application; in the description of the present application, "first", "second", etc. are only used for distinguishing each other, rather than indicating their importance and order, etc.
[0020] In the description of the present application, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" shall be understood in a broad sense. For example, it can be a fixed connection, a movable connection, or a detachable connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements, etc. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.
[0021] The embodiment of the present application provides an enterprise growth scale prediction method, and the method includes the following steps: Step 110: Screen enterprise growth indicators and synthesize the enterprise growth indicators into an enterprise growth scale indicator.
[0022] Among them, all data indicators related to enterprise growth can be screened out, and then these indicators are synthesized into a comprehensive enterprise growth scale indicator. The relevant data indicators can be financial indicators and non-financial indicators. The comprehensive enterprise growth scale indicator is a multi-dimensional comprehensive reflection, which can help better predict the trend of enterprise growth scale.
[0023] Step 120: Collect data according to the enterprise growth scale indicator and generate enterprise growth scale data.
[0024] Among them, based on multi-source data such as the historical financial data, market trends, industry data, and macroeconomic indicators of the enterprise, the enterprise growth scale indicator is used as the key feature for constructing the data set and used as the input data for the subsequent model.
[0025] Step 130: Input the enterprise growth data into the trained enterprise growth scale prediction model for prediction to obtain the enterprise growth scale prediction result.
[0026] Exemplarily, in the embodiments of the present application, a convolutional neural network is selected as the deep learning model, and it is obtained by training based on the selected sample data. The sample data is derived from the multi-source data in the foregoing text. During the training process, the enterprise growth scale index is used as the training label of the model.
[0027] When constructing the convolutional neural network model, the number of convolutional kernels is 64 convolutional kernels with a size of 3*1. Adding BatchNormalization helps to accelerate the training process and improve the model stability. The pooling layer is constructed using 2*2 max pooling, and the second convolutional layer and the corresponding pooling layer are repeatedly constructed. Dropout is added before constructing the fully connected layer to prevent overfitting.
[0028] Based on the above steps, by combining the indicators related to enterprise growth into a comprehensive enterprise growth scale index, the trend of the enterprise growth scale is comprehensively evaluated and predicted, thereby improving the accuracy of the enterprise growth scale prediction.
[0029] In some possible implementation manners, step 110 specifically includes the following steps: Step 111: Calculate the difference coefficient of the entropy values of each enterprise growth indicator.
[0030] Among them, the entropy value method is based on the concept of information entropy and is used to measure the dispersion degree of data indicators. The larger its value, the more dispersed the data distribution and the more information it contains.
[0031]
[0032] is the entropy value of the th indicator; is a constant, usually taking , so that the entropy value is between 0 and 1; is the standardized proportion of the th enterprise's th indicator value, and the calculation formula is:
[0033] is the th enterprise's th indicator value; is the total number of enterprises; is the natural logarithm.
[0034] The difference coefficient is used to reflect the difference degree of the indicators and is inversely proportional to the entropy value.
[0035]
[0036] is the The coefficient of variation of each indicator.
[0037] Step 112: Calculate the weights of each enterprise growth indicator through the coefficient of variation.
[0038] Among them, the weight reflects the importance of each indicator in the comprehensive evaluation.
[0039]
[0040] is the weight of the th indicator; is the total number of indicators.
[0041] Step 113: Screen the top three enterprise growth indicators in terms of weight and obtain the enterprise growth scale indicator through weighted summation. According to the calculated weights, select the three indicators with the largest weights as the growth indicators. Weight the selected growth indicators and sum them up to obtain the comprehensive indicator.
[0042]
[0043] is the comprehensive enterprise growth scale indicator of the th enterprise; indicates selecting the top three indicators in terms of weight.
[0044] In some possible implementation manners, the method further includes the following steps: Step 140: Integrate the enterprise growth scale prediction model into the reinforcement learning framework.
[0045] Among them, in the reinforcement learning environment, the prediction result of the enterprise growth scale prediction model will be used as part of the reward signal or the state.
[0046] Step 141: Obtain the corresponding enterprise growth scale prediction result by adjusting the investment strategy multiple times in the reinforcement learning framework.
[0047] The reinforcement learning agent conducts multiple trials (referred to as "episodes") in the environment and tries different investment strategies at each time step. At each time step, the enterprise growth prediction model will provide a prediction of the enterprise growth scale based on the current enterprise state and the adopted investment strategy.
[0048] Step 142: Compare the enterprise growth scale prediction results to obtain the optimal investment strategy.
[0049] The goal of the reinforcement learning agent is to find an investment strategy that maximizes long-term rewards through continuous trial-and-error learning. By comparing the predicted results of enterprise growth scale under different strategies, the agent can learn which strategies are more likely to bring about enterprise growth and gradually optimize its investment decisions.
[0050] In some possible implementation manners, the state of the reinforcement learning framework is composed of a set of key feature variables; among them, a set of key variable features includes market growth rate, enterprise profitability, economic indicators, and historical predicted results of enterprise growth scale; Adjusting the investment strategy is a set of investment actions; among them, the investment actions include percentages of increasing, decreasing, or maintaining investments in different assets.
[0051] Taking the key feature variables as state parameters, the investment decision can include how much investment to increase in a certain category, how much investment to decrease in a certain category, or remain unchanged, etc. By adjusting the state parameters through actions such as how much investment to increase in a certain category, how much investment to decrease in a certain category, or remain unchanged, the action affects the state.
[0052] State Is composed of a set of key feature variables, and these variables affect investment decisions. For example:
[0053] Among them, Represents the th key feature variable, which may include but is not limited to market growth rate, company profitability, economic indicators, etc.
[0054] Action Represents an investment decision, which can be the percentage of increasing, decreasing, or maintaining the investment in a certain category. Mathematically expressed as:
[0055] Among them, Is the investment adjustment for the th category of assets, and can be in the following forms:
[0056] Is an adjustment coefficient (which can be a positive number, a negative number, or zero), Is the current investment amount in the th category of assets.
[0057] In some possible implementation manners, the method further includes the following steps: Step 100: Preprocess the attribute data of the enterprise; including data cleaning, screening key feature variables, and data integration.
[0058] Among them, data cleaning includes removing outliers and dealing with missing values by estimating missing values through the method of random forest regression. Feature engineering is used to screen key feature variables. First, each variable in the data is binned by WOE, and then the corresponding IV value is calculated. The data with an IV value less than 0.02 is deleted. For two variables with a correlation coefficient greater than 0.6, the variable with the lower IV value is deleted. Then, all key feature variables are normalized. The Pearson correlation coefficient is used to analyze the correlation between variables. Data integration is to merge data from different sources to form a unified data set.
[0059] The embodiment of the present application also provides an enterprise growth scale prediction system, which includes: An index screening unit 210, configured to screen enterprise growth indexes and synthesize the enterprise growth indexes into an enterprise growth scale index; A data acquisition unit 220, configured to collect data according to the enterprise growth scale index and generate enterprise growth scale data; A growth prediction unit 230, configured to input the enterprise growth data into a trained enterprise growth scale prediction model for prediction to obtain an enterprise growth scale prediction result.
[0060] In some possible implementation manners, The index screening unit 210 is configured to calculate the difference coefficient of the entropy values of each enterprise growth index; The index screening unit 210 is further configured to calculate the weight of each enterprise growth index through the difference coefficient; The index screening unit 210 is further configured to screen the top three enterprise growth indexes with weights and obtain the enterprise growth scale index through weighted summation.
[0061] In some possible implementation manners, the system further includes a reinforcement learning unit 240; The reinforcement learning unit 240 is configured to integrate the enterprise growth scale prediction model into the reinforcement learning framework; The reinforcement learning unit 240 is further configured to obtain the corresponding enterprise growth scale prediction result by adjusting the investment strategy multiple times in the reinforcement learning framework; The reinforcement learning unit 240 is further configured to compare the enterprise growth scale prediction results to obtain the optimal investment strategy.
[0062] In some possible implementation manners, The state of the reinforcement learning framework is composed of a set of key feature variables; among them, a set of key variable features includes market growth rate, enterprise profitability, economic indicators, and historical enterprise growth scale prediction results; Adjusting the investment strategy is a set of investment actions; among them, the investment actions include multiple percentages of increasing, decreasing, or maintaining investments in different assets.
[0063] In some possible embodiments, the system further includes a data processing unit 250; The data processing unit 250 is configured to preprocess the enterprise's attribute data; including data cleaning, screening key feature variables, and data integration.
[0064] The embodiment of the present application further provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer is enabled to execute the method of any possible implementation manner in the above-mentioned various embodiments.
[0065] The embodiment of the present application further provides a computer program product, which includes: computer program code. When the computer program code runs on a computer, the computer is enabled to execute the method of any possible implementation manner in the first aspect or the second aspect.
[0066] The embodiment of the present application further provides a computer program. When the computer program code runs on a computer, the computer is enabled to execute the method of any possible implementation manner in any of the foregoing embodiments.
[0067] Those skilled in the art should be able to realize that in the above one or more examples, the functions described in the embodiments of the present application can be implemented by hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. The computer-readable medium includes a computer storage medium and a communication medium, where the communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.
[0068] Note that the above is only the preferred embodiment of the present application and the applied technical principle. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments, and substitutions can be made by those skilled in the art without departing from the protection scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments. Without departing from the concept of the present application, more other equivalent embodiments can be included, and the scope of the present application is determined by the scope of the appended claims.
Claims
1. A method for predicting the growth scale of an enterprise, characterized in that, The method includes: Screening enterprise growth indicators and synthesizing the enterprise growth indicators into an enterprise growth scale indicator; Collecting data according to the enterprise growth scale indicator to generate enterprise growth scale data; Inputting the enterprise growth data into a trained enterprise growth scale prediction model for prediction to obtain an enterprise growth scale prediction result.
2. The enterprise growth scale prediction method according to claim 1, characterized in that, The screening of enterprise growth indicators and synthesizing the enterprise growth indicators into an enterprise growth scale indicator includes: Calculating the difference coefficient of the entropy values of each enterprise growth indicator; Calculating the weight of each enterprise growth indicator through the difference coefficient; Screening the top three enterprise growth indicators by weight and obtaining the enterprise growth scale indicator through weighted summation.
3. The enterprise growth scale prediction method according to claim 1, characterized in that The method includes: Integrating the enterprise growth scale prediction model into a reinforcement learning framework; Obtaining corresponding enterprise growth scale prediction results by adjusting the investment strategy multiple times in the reinforcement learning framework; Comparing the enterprise growth scale prediction results to obtain the optimal investment strategy.
4. The enterprise growth scale prediction method according to claim 3, wherein: The state of the reinforcement learning framework is composed of a set of key feature variables; among them, the set of key variable features includes market growth rate, enterprise profitability, economic indicators, and historical enterprise growth scale prediction results; The adjustment of the investment strategy is a set of investment actions; among them, the investment actions include multiple percentages of increasing, decreasing, or maintaining investments in different assets.
5. The enterprise growth scale prediction method according to claim 4, characterized in that The method further includes: Preprocessing the attribute data of the enterprise; including data cleaning, screening key feature variables, and data integration.
6. An enterprise growth scale prediction system, characterized in that, The system includes: An indicator screening unit for screening enterprise growth indicators and synthesizing the enterprise growth indicators into an enterprise growth scale indicator; A data collection unit for collecting data according to the enterprise growth scale indicator to generate enterprise growth scale data; A growth prediction unit for inputting the enterprise growth data into a trained enterprise growth scale prediction model for prediction to obtain an enterprise growth scale prediction result.
7. The enterprise growth scale prediction system according to claim 6, wherein: The indicator screening unit is used to calculate the difference coefficient of the entropy values of each enterprise growth indicator; The indicator screening unit is further used to calculate the weight of each enterprise growth indicator through the difference coefficient; The indicator screening unit is further used to screen the top three enterprise growth indicators by weight and obtain the enterprise growth scale indicator through weighted summation.
8. The enterprise growth scale prediction system according to claim 6, characterized in that: The system further includes a reinforcement learning unit; The reinforcement learning unit is used to integrate the enterprise growth scale prediction model into a reinforcement learning framework; The reinforcement learning unit is further used to obtain corresponding enterprise growth scale prediction results by adjusting the investment strategy multiple times in the reinforcement learning framework; The reinforcement learning unit is further used to compare the enterprise growth scale prediction results to obtain the optimal investment strategy.
9. The enterprise growth scale prediction system according to claim 8, wherein: The state of the reinforcement learning framework is composed of a set of key feature variables; among them, the set of key variable features includes market growth rate, enterprise profitability, economic indicators, and historical enterprise growth scale prediction results; The adjusted investment strategy is a set of investment actions; among them, the investment actions include multiple percentages of increasing, decreasing, or maintaining investments in different assets.
10. The enterprise growth scale prediction system according to claim 9, characterized in that: The system further includes a data processing unit; The data processing unit is used for preprocessing the attribute data of the enterprise; including data cleaning, screening key feature variables, and data integration.