Quantitative evaluation method, device and equipment for subject operation data, storage medium and program product

Abstract

The invention relates to a quantitative evaluation method and device for subject operation data, equipment, a storage medium and a program product. The method comprises the following steps: screening out comparable reference subjects; fitting industry operation data through a unary regression model based on all reference subjects of the industry to which the comparable reference subject belongs; calculating an average income growth rate and an average income level of the industry; estimating risk parameters of the comparable reference subject based on a unary regression model, and calculating a right value quantized value of the comparable reference subject; calculating an operating value quantized value of the comparable reference subject by combining liability data and cash holding data of the comparable reference subject; respectively calculating the ratio of a preset number of comparable reference subjects to the financial parameters corresponding to the subject to be evaluated to obtain financial parameter adjustment weights; value adjustment parameters are obtained through weighted summation calculation, and the operation value quantification interval of the subject to be evaluated is determined. The method can improve the objectivity and practicability of the evaluation of the operating value of the subject who is not listed or lacks public market data.

Description

Technical Field

[0001] This application relates to the field of company valuation technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for the quantitative evaluation of a main business data. Background Technology

[0002] With the development of corporate valuation techniques, two major valuation methods have emerged: relative valuation and absolute valuation. Relative valuation determines a company's value based on its relative price. This method compares a company with other companies in the same industry or related fields, determining its relative value by comparing its market price, yield, financial parameters, etc. Absolute valuation, on the other hand, discounts a company's expected future cash flows using future discounting methods to reveal the intrinsic value of the company's stock and thus calculate the company's overall value.

[0003] However, absolute valuation methods rely heavily on expectations of a company's future parameters, such as growth rate and depreciation rate. These indicators are subject to significant negotiation for non-listed companies, leading to substantial differences and subjectivity in valuation results, making it difficult to achieve fair value. Relative valuation methods, on the other hand, require valuation personnel to have industry knowledge to determine comparable companies. Furthermore, companies may operate across multiple industries, making manually selected comparable companies potentially unrepresentative and leading to unfair valuation results. Summary of the Invention

[0004] Therefore, it is necessary to provide a quantitative evaluation method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can provide fair and equitable business data to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for quantitatively evaluating the operating data of a business entity, including:

[0006] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0007] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0008] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0009] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0010] In one embodiment, a predetermined number of benchmark subjects are selected as comparable reference subjects, including:

[0011] Obtain the main business description text, patent technology abstract text, and product feature description text of the entity to be evaluated and the benchmark entity;

[0012] The text is preprocessed by word segmentation and stop word removal using a processing model based on the RoFormer architecture to generate sentence vectors.

[0013] The similarity between the main business description text of the entity to be evaluated and each benchmark entity, the similarity between the patent technology abstract text and the product feature description text are calculated in sequence. Based on the similarity ranking results, a preset number of benchmark entities are selected as comparable reference entities.

[0014] In one embodiment, before generating sentence vectors after preprocessing each text by word segmentation and stop word removal using a RoFormer-based processing model, the following steps are also included:

[0015] After preprocessing the text, it is input into the processing model based on the RoFormer architecture and the SimBERT model respectively to obtain the target sentence vector and the reference sentence vector.

[0016] A sentence vector optimization loss function is introduced and minimized through backpropagation to optimize the parameters in the processing model based on the RoFormer architecture, so that the similarity distribution between the target sentence vector and the reference sentence vector remains consistent.

[0017] In one embodiment, based on the similarity ranking results, a predetermined number of benchmark subjects are selected as comparable reference subjects, including:

[0018] The similarity between the main business description text of the subject to be evaluated and each benchmark subject is sorted from largest to smallest, and the benchmark subjects with the first preset proportion are selected to form the initial reference subject pool.

[0019] The similarity between the patent technology abstract texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top two preset proportions of benchmark subjects are selected; the similarity between the product feature description texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top three preset proportions of benchmark subjects are selected; the intersection of the two selected benchmark subjects is taken to obtain the final reference subject pool.

[0020] Based on the similarity between the patent technology abstract text and the product feature description text between the subject to be evaluated and each benchmark subject in the final reference subject pool, the comprehensive similarity between the subject to be evaluated and each benchmark subject in the final reference subject pool is calculated; based on the comprehensive similarity ranking results, a predetermined number of benchmark subjects are selected as comparable reference subjects.

[0021] In one embodiment, a preset number of core industry-specific financial parameters are selected, including:

[0022] The benchmark entities in the same industry as the comparable reference entities are marked as target samples, and the remaining benchmark entities are marked as non-target samples. The financial statement data of all benchmark entities are binned, and the weighted evidence value of each bin is calculated based on the proportion of target samples in each bin to the total number of target samples, and the proportion of non-target samples to the total number of non-target samples. Based on the weighted evidence value of each bin, the information value of each bin is calculated. The information values ​​of each bin are summed to obtain the total information value of a single financial parameter, and a preset number of core industry characteristic financial parameters are selected based on the total information value.

[0023] In one embodiment, based on the value adjustment parameters and operating value quantification values ​​corresponding to each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined, including:

[0024] Multiply the quantitative value of the operating value of each comparable reference entity by the corresponding value adjustment parameter to obtain the quantitative value of the operating value of the entity to be evaluated for a single comparable reference entity;

[0025] The operating value quantification values ​​of the entity to be evaluated are sorted according to the comparable reference entities to obtain the minimum and maximum values. The minimum value is used as the lower limit of the interval and the maximum value is used as the upper limit of the interval to obtain the operating value quantification range of the entity to be evaluated.

[0026] Secondly, this application also provides a device for quantitatively evaluating the main operating data, including:

[0027] The filtering module is used to select a preset number of benchmark subjects as comparable reference subjects for the subject to be evaluated;

[0028] The calculation module is used to fit industry operating data to all benchmark entities in the same industry as the comparable entity using a univariate regression model; calculate the industry's average revenue growth rate and average revenue level using the industry operating data; estimate the risk parameters of the comparable entity based on the univariate regression model, and calculate the expected return parameters of the comparable entity by combining the risk-free return parameter and the market return parameter; substitute the average revenue growth rate, average return level, and expected return parameters into the income discount model to calculate the quantified equity value of the comparable entity; and calculate the quantified operating value of the comparable entity by combining the liability data and cash holding data of the comparable entity.

[0029] The calculation module is also used to filter out a preset number of core industry financial parameters, and based on the preset number of core industry financial parameters, calculate the ratio of the corresponding financial parameters of a preset number of comparable reference entities to the entity to be evaluated, and obtain the financial parameter adjustment weights; assign influence coefficients to each financial parameter adjustment weight, and obtain the value adjustment parameters by weighted summation.

[0030] The assessment module is used to determine the quantitative range of the operating value of the entity to be assessed based on the value adjustment parameters and the quantitative value of operating value corresponding to each comparable reference entity.

[0031] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0032] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0033] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0034] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0035] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0036] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0037] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0038] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0039] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0040] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0041] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0042] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0043] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0044] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0045] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0046] The aforementioned quantitative assessment methods, devices, computer equipment, computer-readable storage media, and computer program products for the main entity's operating data firstly, use all benchmark entities in the industry as samples to fit a univariate regression model, calculate the expected return parameters of comparable reference entities by combining parameters such as risk-free return and market return, and then substitute them into a discounted cash flow model to calculate equity value and combine it with liability and cash data to derive operating value. This overcomes the estimation bias caused by a single reference entity or partial industry data, and solidifies the foundation for value quantification. Secondly, it screens the core characteristic financial parameters of the industry and calculates the parameter ratio between the entity to be assessed and comparable reference entities. By allocating influence coefficients for weighting, it obtains value adjustment parameters, transforming individual differences in financial characteristics into quantifiable adjustment basis, and solving the problem of insufficient adaptability of financial dimensions between different entities. Finally, based on the value adjustment parameters and the operating value of comparable reference entities, it determines the quantitative range of the operating value of the entity to be assessed, rather than a single value. This ensures both the accuracy of the assessment and covers a reasonable range of valuation fluctuations through the range form, effectively reducing the valuation risk caused by subjective assumptions, and improving the objectivity and practicality of operating value assessment for non-listed entities or entities lacking public market data. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is an application environment diagram of a quantitative evaluation method for main operating data in one embodiment;

[0049] Figure 2 This is a flowchart illustrating a method for quantitatively evaluating the main operating data in one embodiment;

[0050] Figure 3 This is a diagram illustrating the overall architecture of a quantitative evaluation method for main operating data in one embodiment.

[0051] Figure 4 This is a flowchart illustrating step 1 of a quantitative evaluation method for main operating data in one embodiment;

[0052] Figure 5 This is a flowchart illustrating step 3 of the quantitative evaluation method for main operating data in one embodiment;

[0053] Figure 6 This is a flowchart illustrating step 4 of the quantitative evaluation method for main operating data in one embodiment;

[0054] Figure 7 This is a structural block diagram of a device for quantitatively evaluating the main operating data in one embodiment;

[0055] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0057] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0058] The quantitative evaluation method for main operating data provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server. First, a univariate regression model is fitted using all benchmark entities in the industry as samples. Parameters such as risk-free return and market return are combined to calculate the expected return parameters of comparable entities. These are then substituted into a discounted cash flow model to calculate equity value, and combined with debt and cash data to derive operating value. This approach overcomes estimation biases caused by single benchmark entities or partial industry data, solidifying the foundation for value quantification. Second, core industry-specific financial parameters are selected, and the parameter ratios between the entity to be evaluated and comparable entities are calculated. Value adjustment parameters are obtained by weighting these parameters using influence coefficients, transforming individual differences in financial characteristics into quantifiable adjustment bases and resolving the issue of insufficient compatibility of financial dimensions among different entities. Finally, based on the value adjustment parameters and the operating value of comparable entities, a quantitative range for the operating value of the entity to be evaluated is determined, rather than a single numerical value. This approach balances the accuracy of the evaluation with a reasonable range covering valuation fluctuations, effectively reducing valuation risks arising from subjective assumptions and comprehensively improving the objectivity and practicality of operating value assessment for unlisted entities or those lacking public market data.

[0059] Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0060] In one exemplary embodiment, such as Figure 2 As shown, a quantitative evaluation method for main business data is provided, which can be applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 202 to 208. Wherein:

[0061] Step 202: For the subject to be evaluated, select a predetermined number of benchmark subjects as comparable reference subjects;

[0062] The entity to be evaluated is a non-listed company, while the benchmark entity is a listed company.

[0063] Step 204: Based on all benchmark entities in the industry to which the comparable entity belongs, fit the industry operating data using a univariate regression model; calculate the industry's average revenue growth rate and average revenue level using the industry operating data; estimate the risk parameters of the comparable entity based on the univariate regression model, and calculate the expected revenue parameters of the comparable entity by combining the risk-free return parameter and the market return parameter; substitute the average revenue growth rate, average revenue level, and expected revenue parameters into the income discount model to calculate the quantified equity value of the comparable entity; and calculate the quantified operating value of the comparable entity by combining the liability data and cash holding data of the comparable entity.

[0064] The univariate regression model is a linear regression analysis model that contains only one independent variable and one dependent variable. Its core is to fit the linear relationship between the independent variable and the dependent variable through mathematical statistics.

[0065] Specifically, using the operating data of all benchmark entities in the same industry as the comparable entity as a sample, an objective industry operating data pattern is fitted through a univariate regression model. Based on this, the average growth rate and average return level of the industry are extracted, providing a benchmark dimension that aligns with industry trends for value assessment. At the same time, the risk parameters of the comparable entity are accurately estimated based on this regression model, and the expected return parameters are calculated by combining risk-free return and market return parameters, thus constructing a quantitative correlation between "risk and return". After substituting the industry return indicators and expected return parameters into the income discount model, the equity value of the comparable entity is obtained, and then the operating value is adjusted by combining the liability and cash holding data.

[0066] Step 206: Select a preset number of core industry financial parameters; based on the preset number of core industry financial parameters, calculate the ratio of the corresponding financial parameters of a preset number of comparable reference entities to the entity to be evaluated, and obtain the financial parameter adjustment weights; assign influence coefficients to each financial parameter adjustment weight, and obtain the value adjustment parameters by weighted summation.

[0067] Among them, the core financial parameters of the industry are the core financial indicators that play a decisive role in the operating value, profitability and risk level of companies in a specific industry, and are the key dimensions for screening the differences between comparable entities and the entity to be evaluated.

[0068] Step 208: Based on the value adjustment parameters and operating value quantification values ​​corresponding to each comparable reference entity, determine the operating value quantification range of the entity to be evaluated.

[0069] Among them, the operating value quantification range is a reasonable range of the operating value of the entity to be evaluated, which is based on the value adjustment parameters of all comparable reference entities, and is the final output of the valuation process.

[0070] The quantitative assessment method for the aforementioned entity's operating data firstly involves fitting a univariate regression model using all benchmark entities in the industry as samples. This model calculates the expected return parameters of comparable entities by combining parameters such as risk-free return and market return. These parameters are then substituted into a discounted cash flow model to calculate equity value, and combined with liability and cash data to derive operating value. This approach overcomes estimation biases caused by single benchmark entities or partial industry data, solidifying the foundation for value quantification. Secondly, it screens core industry-specific financial parameters and calculates the parameter ratios between the entity being assessed and comparable entities. Value adjustment parameters are obtained by weighting these parameters using influence coefficients, transforming individual differences in financial characteristics into quantifiable adjustment bases and resolving the issue of insufficient compatibility of financial dimensions among different entities. Finally, it determines the quantitative range of the entity's operating value based on the value adjustment parameters and the operating value of comparable entities, rather than using a single numerical value. This approach balances the accuracy of the assessment with a reasonable range covering valuation fluctuations, effectively reducing valuation risks arising from subjective assumptions and enhancing the objectivity and practicality of operating value assessment for unlisted entities or those lacking public market data.

[0071] In one embodiment, a predetermined number of benchmark subjects are selected as comparable reference subjects, including:

[0072] Obtain the main business description text, patent technology abstract text, and product feature description text of the entity to be evaluated and the benchmark entity;

[0073] The text is preprocessed by word segmentation and stop word removal using a processing model based on the RoFormer architecture to generate sentence vectors.

[0074] The similarity between the main business description text of the entity to be evaluated and each benchmark entity, the similarity between the patent technology abstract text and the product feature description text are calculated in sequence. Based on the similarity ranking results, a preset number of benchmark entities are selected as comparable reference entities.

[0075] Specifically: 1) Obtain the sentence vectors u and v1, v2, ..., v of the main business scope of the subject to be evaluated and the benchmark subject through the RoFormer-Sim model. n Using cosine similarity, text similarity scores are obtained and ranked. The top 20% of benchmark entities are selected as the comparable company pool K1 for the next calculation. The cosine similarity calculation formula is:

[0076]

[0077] Among them, the RoFormer-Sim model is an open-source pre-trained model based on RoFormer. It is based on the UniLM concept, introduces BART-like training, and is obtained through distillation.

[0078] For the same type of text, SimBERT yields sentence vector u, while RoFormer-Sim yields sentence vector v. The loss function (processing model) for the RoFormer-Sim retrieval model is:

[0079]

[0080] Where λ is a non-negative real hyperparameter used to balance the weights of different loss terms.

[0081] 2) Based on the company pool K1, the text data from the patent abstract and product description are compared with the subject to be evaluated. The sentence vectors are obtained through the RoFormer-Sim model to calculate the similarity, and the text similarity SP and SQ are obtained. SP and SQ are the abstract similarity and product description similarity, respectively.

[0082] 3) Select the top 20% of benchmark subjects in terms of text similarity SP and the top 20% in terms of text similarity SQ, and combine them into a set K2 as the final pool of comparable companies.

[0083] 4) Calculate the final similarity S = 0.5 * SP + 0.5 * SQ, sort the S, and select the three benchmark subjects with the largest similarity as comparable reference subjects.

[0084] The above embodiments break through the limitations of traditional methods that rely solely on industry classification or financial data to screen comparable entities. They comprehensively collect core textual information such as main business descriptions, patent technology summaries, and product feature descriptions, covering the core dimensions of company operations and avoiding reference entity bias caused by single-dimensional matching. By leveraging the RoFormer architecture model to perform word segmentation, stop word removal preprocessing, and sentence vector generation, the text is fully utilized to accurately capture text contextual relationships through RoFormer's rotating positional encoding. By calculating text similarity across dimensions and sorting and screening a preset number of benchmark entities, quantitative semantic similarity replaces subjective experience judgment, significantly improving the matching degree between comparable reference entities and the entity to be evaluated at the business, technology, and product levels. This lays a high-quality reference foundation for subsequent quantitative assessment of the entity's operating value and effectively reduces valuation errors caused by improper matching of comparable entities.

[0085] In one embodiment, before generating sentence vectors after preprocessing each text by word segmentation and stop word removal using a RoFormer-based processing model, the following steps are also included:

[0086] After preprocessing the text, it is input into the processing model based on the RoFormer architecture and the SimBERT model respectively to obtain the target sentence vector and the reference sentence vector.

[0087] A sentence vector optimization loss function is introduced and minimized through backpropagation to optimize the parameters in the processing model based on the RoFormer architecture, so that the similarity distribution between the target sentence vector and the reference sentence vector remains consistent.

[0088] The RoFormer-Sim model is used to obtain the sentence vectors u and v1, v2, ..., v of the main business scope of the subject to be evaluated and the benchmark subject. n Using cosine similarity, text similarity scores are obtained and ranked. The top 20% of benchmark entities are selected as the comparable company pool K1 for the next calculation. The cosine similarity calculation formula is:

[0089]

[0090] Among them, the RoFormer-Sim model is an open-source pre-trained model based on RoFormer. It is based on the UniLM concept, introduces BART-like training, and is obtained through distillation.

[0091] For the same type of text, SimBERT yields sentence vector u, while RoFormer-Sim yields sentence vector v. The loss function (processing model) for the RoFormer-Sim retrieval model is:

[0092]

[0093] Where λ is a non-negative real hyperparameter used to balance the weights of different loss terms.

[0094] In the above embodiments, the mature semantic representation capabilities of the SimBERT model provide a reliable optimization reference for the RoFormer architecture processing model. Furthermore, the loss function enables targeted adjustment of parameters, effectively compensating for the limitations of a single model in capturing text semantics. This allows the sentence vectors generated by the optimized RoFormer architecture processing model to more accurately reflect the core semantics of the text, thereby improving the accuracy of subsequent text similarity calculations between the subject to be evaluated and the benchmark subject.

[0095] In one embodiment, based on the similarity ranking results, a predetermined number of benchmark subjects are selected as comparable reference subjects, including:

[0096] The similarity between the main business description text of the subject to be evaluated and each benchmark subject is sorted from largest to smallest, and the benchmark subjects with the first preset proportion are selected to form the initial reference subject pool.

[0097] The similarity between the patent technology abstract texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top two preset proportions of benchmark subjects are selected; the similarity between the product feature description texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top three preset proportions of benchmark subjects are selected; the intersection of the two selected benchmark subjects is taken to obtain the final reference subject pool.

[0098] Based on the similarity between the patent technology abstract text and the product feature description text between the subject to be evaluated and each benchmark subject in the final reference subject pool, the comprehensive similarity between the subject to be evaluated and each benchmark subject in the final reference subject pool is calculated; based on the comprehensive similarity ranking results, a predetermined number of benchmark subjects are selected as comparable reference subjects.

[0099] The RoFormer-Sim model is used to obtain the sentence vectors u and v1, v2, ..., v of the main business scope of the subject to be evaluated and the benchmark subject. n Using cosine similarity, text similarity scores are obtained and ranked. The top 20% of benchmark entities are selected as the comparable company pool K1 for the next calculation. The cosine similarity calculation formula is:

[0100]

[0101] Among them, the RoFormer-Sim model is an open-source pre-trained model based on RoFormer. It is based on the UniLM concept, introduces BART-like training, and is obtained through distillation.

[0102] In the above embodiments, the main business description text similarity is used as the primary dimension for sorting and selecting preceding entities to form an initial reference entity pool, ensuring that the core business direction of comparable entities is highly consistent with that of the entity to be evaluated. Then, within the initial reference entity pool, preceding entities are screened based on the similarity of patent technology abstracts and product characteristic description texts, and the intersection is taken to accurately focus on the dual matching of the two in core technology and product attributes, eliminating interfering entities that only match in a single dimension. The final comparable entities are determined by comprehensive similarity calculation and sorting. This not only improves the screening efficiency by narrowing the scope layer by layer, but also strengthens the matching accuracy by multi-dimensional intersection and comprehensive scoring, effectively avoiding the comparable entity bias problem caused by traditional single-dimensional or disordered screening.

[0103] In one embodiment, a preset number of core industry-specific financial parameters are selected, including:

[0104] The benchmark entities in the same industry as the comparable reference entities are marked as target samples, and the remaining benchmark entities are marked as non-target samples. The financial statement data of all benchmark entities are binned, and the weighted evidence value of each bin is calculated based on the proportion of target samples in each bin to the total number of target samples, and the proportion of non-target samples to the total number of non-target samples. Based on the weighted evidence value of each bin, the information value of each bin is calculated. The information values ​​of each bin are summed to obtain the total information value of a single financial parameter, and a preset number of core industry characteristic financial parameters are selected based on the total information value.

[0105] All companies in the Shenwan industry where the comparable reference entity is located are marked as 1, and other companies are marked as 0. After binning the three major financial statements of the comparable reference entity, the weighted evidence value (WOE) and information value (IV) are calculated. The six core characteristic financial parameters i1, i2, ..., i6 with the largest IV values ​​are selected as the most influential adjustment coefficient indicators for the industry.

[0106] The formulas for calculating WOE and IV values ​​are as follows:

[0107]

[0108]

[0109]

[0110] Where Pyi represents the proportion of sample company labeled 1 in this bin to all companies labeled 1, and Pni represents the proportion of sample company labeled 0 in this bin to all companies labeled 0.

[0111] In the above embodiments, the weighted evidence value can accurately quantify the ability of binned data to distinguish industry attributes, while the information value further measures the representational effectiveness of financial parameters on industry characteristics. The ranking and screening of total information value ensures that the selected parameters are all key indicators that have a significant impact on the industry's operating value, effectively eliminating redundant or low-correlation parameters, and providing a high-quality core dimension for subsequent quantitative adjustment of the financial differences between the entity to be evaluated and comparable entities, thus ensuring the accuracy of value adjustment from the parameter screening level.

[0112] In one embodiment, based on the value adjustment parameters and operating value quantification values ​​corresponding to each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined, including:

[0113] Multiply the quantitative value of the operating value of each comparable reference entity by the corresponding value adjustment parameter to obtain the quantitative value of the operating value of the entity to be evaluated for a single comparable reference entity;

[0114] The operating value quantification values ​​of the entity to be evaluated are sorted according to the comparable reference entities to obtain the minimum and maximum values. The minimum value is used as the lower limit of the interval and the maximum value is used as the upper limit of the interval to obtain the operating value quantification range of the entity to be evaluated.

[0115] Specifically: 1) Based on the core characteristic financial parameters i1, i2, ..., i6 of the industry, calculate the adjusted weights n1, ..., n6 of the financial parameters corresponding to the three comparable entities and the entity to be evaluated. Taking one of the comparable entities as an example, its core characteristic financial parameters and the core characteristic financial parameters of the entity to be evaluated are j1, ..., j6 and l1, ..., l6 respectively, then n i =l i / j i .

[0116] 2) Adjust the weights of the obtained financial parameters according to their influence w1, ..., w6. The appraiser may make their own judgment based on industry experience. It is assumed that the influence of each coefficient is the same, but it must meet the following requirements. .

[0117] 3) Calculate the adjustment coefficient for the final comparable benchmark operating value quantification. Adjustments are made based on the quantified value V of the business, resulting in Q=Z*V.

[0118] 4) Take the sum of the quantitative operating value values ​​of the above three comparable entities to obtain the final quantitative operating value range.

[0119] In the above embodiments, the quantified operating value of each comparable entity is multiplied by its exclusive value adjustment parameter, ensuring that the value of each comparable entity accurately matches the financial characteristics of the entity being evaluated, resulting in an individual quantified reference value that fits the attributes of the entity being evaluated. Then, by ranking these reference values ​​and taking the extreme values, a value range is defined. This approach not only covers the reasonable fluctuation range of the entity's value by utilizing the adjusted values ​​of multiple comparable entities, but also outputs the results in a clear manner with a "minimum value as the lower limit and the maximum value as the upper limit," avoiding the one-sidedness and subjectivity of a single numerical valuation. This effectively builds upon the results of the previous quantitative adjustment of financial differences, solving the problem of accurate value measurement caused by the lack of public market pricing for non-listed companies. The output quantified range provides clear reference boundaries for evaluation decisions while retaining flexibility in practice, ensuring the reliability and practicality of the operating value assessment at the final result level.

[0120] In one embodiment, reference Figure 3 The figure shows a quantitative evaluation method for the main operating data in a specific embodiment.

[0121] Step 1: Reference Figure 4As shown, textual data such as the main business scope, patents, and product descriptions of the benchmark entity (listed company) and the entity to be evaluated (non-listed company) are obtained. The processed textual data is then used to generate text vectors through the RoFormer-Sim processing model, and the text similarity is calculated. The benchmark entity with higher similarity is selected as the comparable reference entity.

[0122] 1) Obtain the sentence vectors u and v1, v2, ..., v of the main business scope of the subject to be evaluated and the benchmark subject using the RoFormer-Sim model. n Using cosine similarity, text similarity scores are obtained and ranked. The top 20% of benchmark entities are selected as the comparable company pool K1 for the next calculation. The cosine similarity calculation formula is:

[0123] (1)

[0124] Among them, the RoFormer-Sim model is an open-source pre-trained model based on RoFormer. It is based on the UniLM concept, introduces BART-like training, and is obtained through distillation.

[0125] For the same type of text, SimBERT yields sentence vector u, while RoFormer-Sim yields sentence vector v. The loss function (processing model) for the RoFormer-Sim retrieval model is:

[0126] (2)

[0127] Where λ is a non-negative real hyperparameter used to balance the weights of different loss terms.

[0128] 2) Based on the company pool K1, the text data from the patent abstract and product description are compared with the subject to be evaluated. The sentence vectors are obtained through the RoFormer-Sim model to calculate the similarity, and the text similarity SP and SQ are obtained. SP and SQ are the abstract similarity and product description similarity, respectively.

[0129] 3) Select the top 20% of benchmark subjects in terms of text similarity SP and the top 20% in terms of text similarity SQ, and combine them into a set K2 as the final pool of comparable companies.

[0130] 4) Calculate the final similarity S = 0.5 * SP + 0.5 * SQ, sort the S, and select the three benchmark subjects with the largest similarity as comparable reference subjects.

[0131] Step 2: Reference Figure 5As shown, comparable entities are valued using absolute valuation methods such as the discounted cash flow (DCF) model, which represents future cash flows. Here, the DDM dividend discount model is used as an example: necessary parameters such as dividends, yield, growth rate, and risk coefficient of the comparable entity are obtained to calculate the quantitative value of the comparable entity's operating value.

[0132] 1) Select all companies in the Shenwan industry where comparable companies are located, and use a univariate regression model to calculate the industry’s average dividend growth rate g and the company’s recent average dividend level D0.

[0133] 2) Using a univariate regression model, the risk coefficient β of comparable entities is estimated, and government bonds are selected as the risk-free rate r. f The average growth rate of the Shanghai Composite Index is used as the market return r. m Substitute the values ​​into the following formula to calculate the expected rate of return r of the comparable reference subject.

[0134] (3)

[0135] 3) Substitute the parameters obtained from the above two steps—average dividend level D0, average dividend growth rate g, and expected rate of return r—into the following formula, i.e., the discounted cash flow model, to obtain the estimated equity value of the listed company, E0.

[0136] (4)

[0137] 4) Obtain the liabilities D and cash C of the current comparable entity, and substitute them into the following formula to obtain the final operating value quantification V of the comparable entity.

[0138] (5)

[0139] Step 3: Mark all companies in the Shenwan industry where the comparable reference entity is located as 1, and mark other companies as 0. After binning the three major financial statements of the comparable reference entity, calculate the weighted evidence value (WOE) and information value (IV). Select the six core characteristic financial parameters i1, i2, ..., i6 with the largest IV values ​​as the most influential adjustment coefficient indicators for the industry.

[0140] The formulas for calculating WOE and IV values ​​are as follows:

[0141] (6)

[0142] (7)

[0143] (8)

[0144] Where Pyi represents the proportion of sample company labeled 1 in this bin to all companies labeled 1, and Pni represents the proportion of sample company labeled 0 in this bin to all companies labeled 0.

[0145] Step 4: Reference Figure 6 As shown, based on the six core industry financial parameters obtained in step 3 and the corresponding financial parameters of the entity to be evaluated, the financial parameter adjustment weights are obtained, and the quantitative value of the comparable reference entity obtained in step 2 is adjusted to obtain the quantitative range of the entity's operating value.

[0146] 1) Based on the core characteristic financial parameters i1, i2, ..., i6 of the industry, calculate the adjusted weights n1, ..., n6 of the financial parameters corresponding to the three comparable entities and the entity to be evaluated. Taking one of the comparable entities as an example, its core characteristic financial parameters and those of the entity to be evaluated are j1, ..., j6 and l1, ..., l6 respectively, then n... i =l i / j i .

[0147] 2) Adjust the weights of the obtained financial parameters according to their influence w1, ..., w6. The appraiser may make their own judgment based on industry experience. It is assumed that the influence of each coefficient is the same, but it must meet the following requirements. .

[0148] 3) Calculate the adjustment coefficient for the final comparable benchmark operating value quantification. Adjustments are made based on the quantified value V of the business, resulting in Q=Z*V.

[0149] 4) Take the sum of the quantitative operating value values ​​of the three comparable entities in the above steps to obtain the final quantitative operating value range.

[0150] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0151] Based on the same inventive concept, this application also provides a device for quantitatively evaluating main operating data to implement the above-mentioned method for quantitatively evaluating main operating data. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the device for quantitatively evaluating main operating data provided below can be found in the limitations of the method for quantitatively evaluating main operating data described above, and will not be repeated here.

[0152] In one exemplary embodiment, such as Figure 7 As shown, a quantitative evaluation device for main business data is provided, including: a screening module 702, a calculation module 704, and an evaluation module 706, wherein:

[0153] The screening module 702 is used to screen out a preset number of benchmark subjects as comparable reference subjects for the subject to be evaluated;

[0154] Calculation module 704 is used to fit industry operating data to all benchmark entities in the same industry as the comparable entity using a univariate regression model; calculate the industry's average revenue growth rate and average revenue level using the industry operating data; estimate the risk parameters of the comparable entity based on the univariate regression model, and calculate the expected return parameters of the comparable entity by combining the risk-free return parameter and the market return parameter; substitute the average revenue growth rate, average return level, and expected return parameters into the income discount model to calculate the quantified equity value of the comparable entity; and calculate the quantified operating value of the comparable entity by combining the liability data and cash holding data of the comparable entity.

[0155] The calculation module 704 is also used to filter out a preset number of industry core characteristic financial parameters, and based on the preset number of industry core characteristic financial parameters, calculate the ratio of the corresponding financial parameters of a preset number of comparable reference subjects to the subject to be evaluated, and obtain the financial parameter adjustment weights; assign influence coefficients to each financial parameter adjustment weight, and obtain the value adjustment parameters by weighted summation;

[0156] The assessment module 706 is used to determine the quantitative range of the operating value of the entity to be assessed based on the value adjustment parameters and the quantitative value of operating value corresponding to each comparable reference entity.

[0157] In one embodiment, the filtering module 702 is used to filter out a preset number of benchmark subjects as comparable reference subjects, including:

[0158] Obtain the main business description text, patent technology abstract text, and product feature description text of the entity to be evaluated and the benchmark entity;

[0159] The text is preprocessed by word segmentation and stop word removal using a processing model based on the RoFormer architecture to generate sentence vectors.

[0160] The similarity between the main business description text of the entity to be evaluated and each benchmark entity, the similarity between the patent technology abstract text and the product feature description text are calculated in sequence. Based on the similarity ranking results, a preset number of benchmark entities are selected as comparable reference entities.

[0161] In one embodiment, the filtering module 702 is further configured to, before generating sentence vectors after preprocessing each text by word segmentation and stop word removal using a processing model based on the RoFormer architecture, include:

[0162] After preprocessing the text, it is input into the processing model based on the RoFormer architecture and the SimBERT model respectively to obtain the target sentence vector and the reference sentence vector.

[0163] A sentence vector optimization loss function is introduced and minimized through backpropagation to optimize the parameters in the processing model based on the RoFormer architecture, so that the similarity distribution between the target sentence vector and the reference sentence vector remains consistent.

[0164] In one embodiment, the filtering module 702 is further configured to filter out a preset number of benchmark subjects as comparable reference subjects based on the similarity ranking results, including:

[0165] The similarity between the main business description text of the subject to be evaluated and each benchmark subject is sorted from largest to smallest, and the benchmark subjects with the first preset proportion are selected to form the initial reference subject pool.

[0166] The similarity between the patent technology abstract texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top two preset proportions of benchmark subjects are selected; the similarity between the product feature description texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top three preset proportions of benchmark subjects are selected; the intersection of the two selected benchmark subjects is taken to obtain the final reference subject pool.

[0167] Based on the similarity between the patent technology abstract text and the product feature description text between the subject to be evaluated and each benchmark subject in the final reference subject pool, the comprehensive similarity between the subject to be evaluated and each benchmark subject in the final reference subject pool is calculated; based on the comprehensive similarity ranking results, a predetermined number of benchmark subjects are selected as comparable reference subjects.

[0168] In one embodiment, the computing module 704 is further configured to:

[0169] The benchmark entities in the same industry as the comparable reference entities are marked as target samples, and the remaining benchmark entities are marked as non-target samples. The financial statement data of all benchmark entities are binned, and the weighted evidence value of each bin is calculated based on the proportion of target samples in each bin to the total number of target samples, and the proportion of non-target samples to the total number of non-target samples. Based on the weighted evidence value of each bin, the information value of each bin is calculated. The information values ​​of each bin are summed to obtain the total information value of a single financial parameter, and a preset number of core industry characteristic financial parameters are selected based on the total information value.

[0170] In one embodiment, the evaluation module 706 is further configured to:

[0171] Multiply the quantitative value of the operating value of each comparable reference entity by the corresponding value adjustment parameter to obtain the quantitative value of the operating value of the entity to be evaluated for a single comparable reference entity;

[0172] The operating value quantification values ​​of the entity to be evaluated are sorted according to the comparable reference entities to obtain the minimum and maximum values. The minimum value is used as the lower limit of the interval and the maximum value is used as the upper limit of the interval to obtain the operating value quantification range of the entity to be evaluated.

[0173] Each module in the aforementioned quantitative evaluation device for main business data can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0174] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a method for quantitatively evaluating core business data. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0175] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0176] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0177] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0178] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0179] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0180] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0181] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0182] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0183] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0184] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0185] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0186] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0187] For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities;

[0188] Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data.

[0189] A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation.

[0190] Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

[0191] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0192] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0193] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0194] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for quantitatively evaluating the main business data, characterized in that, The method includes: For the entity to be evaluated, a predetermined number of benchmark entities are selected as comparable reference entities; Based on all benchmark entities in the same industry, industry operating data is fitted using a univariate regression model. The average revenue growth rate and average return level of the industry are calculated using the industry operating data. Risk parameters of the comparable entities are estimated based on the univariate regression model, and expected return parameters are calculated by combining risk-free return parameters and market return parameters. The average revenue growth rate, average return level, and expected return parameters are substituted into a discounted cash flow model to calculate the quantified equity value of the comparable entities. The operating value of the comparable entities is calculated by combining their liability data and cash holding data. A predetermined number of core industry financial parameters are selected. Based on these parameters, the ratios of the corresponding financial parameters of a predetermined number of comparable entities to the entity being evaluated are calculated to obtain the financial parameter adjustment weights. An influence coefficient is assigned to each financial parameter adjustment weight, and the value adjustment parameters are obtained by weighted summation. Based on the value adjustment parameters and operating value quantification values ​​of each comparable reference entity, the operating value quantification range of the entity to be evaluated is determined.

2. The method according to claim 1, characterized in that, The selection of a preset number of benchmark subjects as comparable reference subjects includes: Obtain the main business description text, patent technology abstract text, and product feature description text of the entity to be evaluated and the benchmark entity; The text is preprocessed by word segmentation and stop word removal using a processing model based on the RoFormer architecture to generate sentence vectors. The similarity between the main business description text of the entity to be evaluated and each benchmark entity, the similarity between the patent technology abstract text and the product feature description text are calculated in sequence. Based on the similarity ranking results, a preset number of benchmark entities are selected as comparable reference entities.

3. The method according to claim 2, characterized in that, Before generating sentence vectors after preprocessing each text by word segmentation and stop word removal using the RoFormer architecture-based processing model, the following steps are also included: After preprocessing the text, it is input into the processing model based on the RoFormer architecture and the SimBERT model respectively to obtain the target sentence vector and the reference sentence vector. A sentence vector optimization loss function is introduced and minimized through backpropagation to optimize the parameters in the processing model based on the RoFormer architecture, so that the similarity distribution between the target sentence vector and the reference sentence vector remains consistent.

4. The method according to claim 2, characterized in that, The step of selecting a predetermined number of benchmark subjects as comparable reference subjects based on similarity ranking results includes: The similarity between the main business description text of the subject to be evaluated and each benchmark subject is sorted from largest to smallest, and the benchmark subjects with the first preset proportion are selected to form the initial reference subject pool. The similarity between the patent technology abstract texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top two preset proportions of benchmark subjects are selected; the similarity between the product feature description texts of the subject to be evaluated and each benchmark subject in the initial reference subject pool is sorted from largest to smallest, and the top three preset proportions of benchmark subjects are selected; the intersection of the two selected benchmark subjects is taken to obtain the final reference subject pool. Based on the similarity between the patent technology abstract text and the product feature description text between the subject to be evaluated and each benchmark subject in the final reference subject pool, the comprehensive similarity between the subject to be evaluated and each benchmark subject in the final reference subject pool is calculated; based on the comprehensive similarity ranking results, a predetermined number of benchmark subjects are selected as comparable reference subjects.

5. The method according to claim 1, characterized in that, The preset number of core industry-specific financial parameters are selected, including: The benchmark entities in the same industry as the comparable reference entities are marked as target samples, and the remaining benchmark entities are marked as non-target samples. The financial statement data of all benchmark entities are binned, and the weighted evidence value of each bin is calculated based on the proportion of target samples in each bin to the total number of target samples, and the proportion of non-target samples to the total number of non-target samples. Based on the weighted evidence value of each bin, the information value of each bin is calculated. The information values ​​of each bin are summed to obtain the total information value of a single financial parameter, and a preset number of core industry characteristic financial parameters are selected based on the total information value.

6. The method according to claim 1, characterized in that, The determination of the quantitative range of the operating value of the entity to be evaluated, based on the value adjustment parameters and quantitative operating value of each comparable reference entity, includes: Multiply the quantitative value of the operating value of each comparable reference entity by the corresponding value adjustment parameter to obtain the quantitative value of the operating value of the entity to be evaluated for a single comparable reference entity; The operating value quantification values ​​of the entity to be evaluated are sorted according to the comparable reference entities to obtain the minimum and maximum values. The minimum value is used as the lower limit of the interval and the maximum value is used as the upper limit of the interval to obtain the operating value quantification range of the entity to be evaluated.

7. A device for quantitatively evaluating main business data, characterized in that, The device includes: The filtering module is used to select a preset number of benchmark subjects as comparable reference subjects for the subject to be evaluated; The calculation module is used to fit industry operating data to all benchmark entities in the same industry as the comparable entity using a univariate regression model; calculate the industry's average revenue growth rate and average revenue level using the industry operating data; estimate the risk parameters of the comparable entity based on the univariate regression model, and calculate the expected return parameters of the comparable entity by combining the risk-free return parameter and the market return parameter; substitute the average revenue growth rate, average return level, and expected return parameters into the income discount model to calculate the quantified equity value of the comparable entity; and calculate the quantified operating value of the comparable entity by combining the liability data and cash holding data of the comparable entity. The calculation module is also used to filter out a preset number of core industry financial parameters, and based on the preset number of core industry financial parameters, calculate the ratio of the corresponding financial parameters of a preset number of comparable reference entities to the entity to be evaluated, and obtain the financial parameter adjustment weights; assign influence coefficients to each financial parameter adjustment weight, and obtain the value adjustment parameters by weighted summation. The assessment module is used to determine the quantitative range of the operating value of the entity to be assessed based on the value adjustment parameters and the quantitative value of operating value corresponding to each comparable reference entity.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images