Manufacturing enterprise scoring method and related device

Abstract

The invention discloses a manufacturing enterprise scoring method and a related device, and relates to the technical field of bank enterprise customer group analysis, and the method comprises the steps: firstly obtaining all types of enterprise information, including fund information, credit information, operation information, qualification information, research and development information and industry and commerce information, of a to-be-scored manufacturing enterprise; extracting feature data of each type of enterprise information; and finally, inputting the feature data of each type of enterprise information into a configured enterprise scoring model to obtain a manufacturing enterprise score output by the enterprise scoring model, the enterprise scoring model being obtained by taking the feature data of each type of enterprise information of a sample manufacturing enterprise as a training sample and taking the score of the sample manufacturing enterprise as a sample label for training. The method has multiple scoring index dimensions and wide data sources, and scores of manufacturing enterprises are more comprehensive, objective and accurate; moreover, the enterprise scoring model is determined by training the machine learning model based on big data, so that the accuracy and credibility of enterprise scoring in the manufacturing industry can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of banking enterprise customer group analysis, and in particular to a manufacturing enterprise scoring method and related devices. Background Art

[0002] The manufacturing industry shoulders the important mission of continuously promoting my country's industries to move towards the middle and high end of the global value chain. Serving advanced manufacturing enterprises is a powerful tool to support the real economy and enhance the resilience of the industrial chain and supply chain. Among the many manufacturing enterprise customers, how to select advanced manufacturing enterprises and accurately reach them is the first threshold to improve the bank's manufacturing service capabilities. Due to the limited ability of account managers, it is difficult to manually summarize the relevant information of various manufacturing enterprises. Even if the account manager obtains the relevant information of multiple manufacturing enterprises, it will be impossible to make a relatively objective and accurate evaluation of the manufacturing enterprises due to subjective factors. Summary of the invention

[0003] To solve the above technical problems, the embodiments of the present application provide a manufacturing enterprise scoring method and related devices to score manufacturing enterprises, thereby selecting advanced manufacturing enterprises and facilitating the creation of financial service solutions for advanced manufacturing enterprises.

[0004] To achieve the above objectives, the present application provides the following technical solutions:

[0005] A manufacturing enterprise scoring method, comprising:

[0006] Obtaining various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information;

[0007] Extract characteristic data of each type of enterprise information;

[0008] The characteristic data of each type of enterprise information is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of sample manufacturing enterprises as sample labels.

[0009] Optionally, each type of enterprise information includes at least two or more sub-types of enterprise information;

[0010] The characteristic data extracted for each type of enterprise information include:

[0011] Extract characteristic data of each sub-type of enterprise information in each type of enterprise information.

[0012] Optionally, the enterprise scoring model includes an input layer, at least one hidden layer and an output layer;

[0013] The input layer is used to perform weighted calculation on the feature data of each sub-type of enterprise information in each type of input enterprise information, wherein the weight of the feature data of each sub-type of enterprise information is a learnable network parameter and is fixed in the model training stage;

[0014] The at least one hidden layer is used to perform feature processing on the result output by the input layer, and send the feature output by the last hidden layer to the output layer;

[0015] The output layer is used to predict the rating of the manufacturing enterprise based on the input features.

[0016] Optionally, there are multiple manufacturing enterprises; the manufacturing enterprise scoring method further includes:

[0017] After obtaining the scores of each of the manufacturing enterprises, the manufacturing enterprises are ranked or divided into different levels according to the scores.

[0018] Optionally, there are multiple manufacturing enterprises; the manufacturing enterprise scoring method further includes:

[0019] After obtaining the scores of each of the manufacturing enterprises, the top n manufacturing enterprises with the highest scores in each manufacturing industry are selected as the leading enterprises of the manufacturing industry for display, where n≥1 and n is an integer.

[0020] Optionally, there are multiple manufacturing enterprises; the manufacturing enterprise scoring method further includes:

[0021] After obtaining the scores of the manufacturing enterprises, the top m manufacturing enterprises with the highest scores in each industrial park are selected as the key enterprises of the industrial park for display, where m≥1 and m is an integer.

[0022] Optionally, there are multiple manufacturing enterprises; the manufacturing enterprise scoring method further includes:

[0023] After obtaining the scores of each of the manufacturing enterprises, the top x manufacturing enterprises with the highest scores in each major project are selected as the leading enterprises of the major project for display, where x≥1 and x is an integer.

[0024] A manufacturing enterprise scoring device, comprising:

[0025] An enterprise information acquisition unit is used to acquire various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information;

[0026] A feature data extraction unit, used to extract feature data of each type of enterprise information;

[0027] The scoring unit is used to input the characteristic data of each type of enterprise information into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of sample manufacturing enterprises as sample labels.

[0028] An electronic device comprises at least one processor and a memory connected to the processor, wherein:

[0029] The memory is used to store computer programs;

[0030] The processor is used to execute the computer program so that the electronic device can implement any one of the manufacturing enterprise scoring methods described above.

[0031] A computer storage medium, wherein the storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement any one of the manufacturing enterprise scoring methods described above.

[0032] Compared with the prior art, the above technical solution has the following advantages:

[0033] The manufacturing enterprise scoring method and related devices provided in the embodiments of the present application first obtain various types of enterprise information of the manufacturing enterprise to be scored, wherein the various types of enterprise information include financial information, credit information, operating information, qualification information, research and development information, and industrial and commercial information; then extract characteristic data of each type of enterprise information; finally, input the characteristic data of each type of enterprise information into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model, so as to facilitate the selection of advanced manufacturing enterprises and create financial service solutions for advanced manufacturing enterprises; wherein the enterprise scoring model is trained using the characteristic data of various types of enterprise information of sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels. The manufacturing enterprise scoring method and related devices provided in the embodiments of the present application score manufacturing enterprises based on multiple types of enterprise information of manufacturing enterprises (including financial information, credit information, operating information, qualification information, research and development information, and industrial and commercial information). The scoring indicators have multiple dimensions and a wide range of data sources, so that the scoring of manufacturing enterprises is more comprehensive, objective, and accurate. In addition, the feature data extracted from each type of enterprise information of the manufacturing enterprises is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the feature data of each type of enterprise information of the sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels. That is, the enterprise scoring model is determined by training a machine learning model based on big data, which is conducive to improving the accuracy and credibility of the scoring of manufacturing enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A flowchart of a method for evaluating an enterprise provided in an embodiment of the present application;

[0036] Figure 2 A structural block diagram of an enterprise scoring device provided in an embodiment of the present application;

[0037] Figure 3 This is a hardware structure block diagram of the electronic device provided in the embodiment of the present application.

[0038] Reference numerals:

[0039] 10-Enterprise information acquisition unit; 20-Feature data extraction unit; 30-Scoring unit. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0042] As mentioned in the background technology section, among the numerous manufacturing enterprise customer groups, how to select advanced manufacturing enterprises and accurately reach them is the first threshold to improve the bank's manufacturing service capabilities. Due to the limited ability of account managers, it is difficult to manually summarize the relevant information of various manufacturing enterprises. Even if the account manager obtains the relevant information of multiple manufacturing enterprises, it will be impossible to make a relatively objective and accurate evaluation of the manufacturing enterprises due to subjective factors.

[0043] In view of this, the present application embodiment provides a manufacturing enterprise scoring method. Figure 1 A schematic diagram of a manufacturing enterprise scoring method provided in an embodiment of the present application is shown. Figure 1 As shown, the scoring method for manufacturing enterprises includes:

[0044] S10: Obtain various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include financial information, credit information, operating information, qualification information, research and development information, and industrial and commercial information.

[0045] S20: Extract characteristic data of each type of enterprise information.

[0046] Among them, the capital information of an enterprise may include the financing information and listing information of the enterprise, which is used to evaluate the capital situation of the enterprise. From the financing information of an enterprise, characteristic data such as the financing round, financing amount, financing method and financing channel of the enterprise can be extracted. From the listing information of an enterprise, characteristic data such as the market share and stock price trend of the enterprise can be extracted.

[0047] The credit information of an enterprise may include its debt information, loan information, etc., which is used to evaluate the credit status of the enterprise. Characteristic data such as the amount of debt and the amount of loan of the enterprise can be extracted from the credit information of the enterprise.

[0048] The business information of an enterprise may include the enterprise's production factor information, supply chain logistics information, and super radar transaction information, etc., which are used to evaluate the enterprise's business situation. The production factors of an enterprise refer to the various resources and conditions required by the enterprise in the production process, which have an important impact on the production efficiency and competitiveness of the enterprise. Specifically, they may include labor, capital, land, and natural resources. The supply chain logistics of an enterprise covers all information related to logistics in the entire process from the procurement of raw materials to the final delivery of products to consumers, which may include procurement and supply, production logistics, sales logistics, and reverse logistics. Super radar usually refers to an advanced data analysis, customer tracking or transaction monitoring tool in an enterprise. It uses big data, artificial intelligence and other technologies to conduct in-depth mining and analysis of various data in the enterprise's transaction process, helping enterprises to discover potential business opportunities, optimize transaction strategies, and improve operational efficiency. From the enterprise's super radar transaction information, characteristic data such as the enterprise's transaction volume, transaction details, and competitors can be extracted.

[0049] The qualification information of an enterprise may include the enterprise's certification information, standard setting information, and government reward and subsidy information, etc., which are used to evaluate the enterprise's qualification level. The enterprise's certification information may include the enterprise's business license, tax registration certificate, organization code certificate, and professional qualification certificate, etc. Among them, the professional qualification certificate is usually issued by the relevant government departments or industry associations to prove that the enterprise has the corresponding technical strength and management level in a specific field. The enterprise's standard setting information may include industry standard setting information and enterprise standard setting information. If the enterprise participates in the formulation of industry standards, it can prove the ability and level of the enterprise in related fields; enterprise standards involve requirements in various aspects of the enterprise's internal technology, management, work, etc., proving the management ability and level of the enterprise. The enterprise's government reward and subsidy information may include various honors received by the enterprise, government funding support, financing support, tax incentives, and project support, etc. It may also include whether the enterprise is a green production and certification enterprise, etc., to explain the importance of enterprise development.

[0050] The R&D information of an enterprise may include the enterprise's R&D funding information, R&D talent information, and intellectual property information, etc., which are used to evaluate the enterprise's R&D level. The enterprise's patent feature data and software copyright feature data can be extracted from the enterprise's intellectual property information. The enterprise's patent feature data may include the number of the enterprise's invention patents, the number of utility model patents, and the number of design patents, and may also include the number of the enterprise's patent families, etc. The software copyright feature data may include the number of the enterprise's software copyrights, etc.

[0051] The business information of an enterprise refers to the various types of information left by the enterprise when it is registered with the industrial and commercial administrative department. This information is the basis for the legal operation of the enterprise and an important way to understand the basic situation of the enterprise and is used to evaluate the compliance of the enterprise. Characteristic data such as the enterprise's entity type, business scope, residence or main business location can be extracted from the enterprise's business information.

[0052] After obtaining various types of enterprise information of manufacturing enterprises, the various types of enterprise information of manufacturing enterprises can be cleaned, which can be done through the Borderline-SMOTE oversampling technology. The Borderline-SMOTE oversampling technology is an oversampling technology used to process class imbalanced data sets, which aims to balance the number of samples of different categories by synthesizing new minority class samples.

[0053] S30: Input the characteristic data of each type of enterprise information into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of the sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels.

[0054] It is understandable that feature data is extracted from each type of enterprise information (including capital information, credit information, business information, qualification information, R&D information, and industrial and commercial information) of the sample manufacturing enterprises as training samples for the enterprise scoring model, and the scores of the sample manufacturing enterprises are used as sample labels of the enterprise scoring model to train the enterprise scoring model. The enterprise scoring model can be a logistic regression model.

[0055] Furthermore, the feature data extracted from each type of enterprise information of the manufacturing enterprise to be scored (including financial information, credit information, operating information, qualification information, R&D information and industrial and commercial information) is input into the trained enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model.

[0056] It can be seen that the manufacturing enterprise scoring method provided in the embodiment of the present application scores manufacturing enterprises based on multiple types of enterprise information of manufacturing enterprises (including financial information, credit information, operating information, qualification information, R&D information and industrial and commercial information). The scoring indicators have multiple dimensions and a wide range of data sources, so that the scoring of manufacturing enterprises is more comprehensive, objective and accurate; and the feature data extracted from each type of enterprise information of the manufacturing enterprises is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model, and the enterprise scoring model is trained by using the feature data of each type of enterprise information of the sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels, that is, the enterprise scoring model is determined by training the machine learning model based on big data, which is conducive to improving the accuracy and credibility of the scoring of manufacturing enterprises.

[0057] Optionally, in some embodiments of the present application, each type of enterprise information includes at least two or more sub-types of enterprise information. For example, the capital information of an enterprise may include the financing information, listing information, etc. of the enterprise; the credit information of an enterprise may include the debt information, loan information, etc. of the enterprise; the business information of an enterprise may include the production factor information, supply chain logistics information, and super radar transaction information of the enterprise; the qualification information of an enterprise may include the certification information, standard setting information, and government reward and subsidy information of the enterprise; the R&D information of an enterprise may include the R&D funds, R&D talents, and intellectual property information of the enterprise; the business information of an enterprise may include the entity type, business scope, residence or main business location information of the enterprise, etc.

[0058] In this embodiment, the feature data extracted from each type of enterprise information in step S20 includes:

[0059] S21: Extracting characteristic data of each sub-type of enterprise information in each type of enterprise information.

[0060] That is to say, when extracting characteristic data from the company's financial information, not only characteristic data such as financing rounds, financing amounts, financing methods and financing channels are extracted from the company's financing information, but also characteristic data such as the company's market share and stock price trends are extracted from the company's listing information.

[0061] When extracting characteristic data from the credit information of an enterprise, characteristic data such as the amount of liabilities of the enterprise is extracted not only from the debt information of the enterprise, but also characteristic data such as the amount of loans of the enterprise is extracted from the loan information of the enterprise.

[0062] When extracting characteristic data from an enterprise's operating information, we can not only extract characteristic data such as the enterprise's labor, capital, land and natural resources from the enterprise's production factor information, but also extract characteristic data such as the enterprise's procurement and supply, production logistics, sales logistics and reverse logistics from the enterprise's supply chain logistics information. We can also extract characteristic data such as the enterprise's transaction volume, transaction details information and competitors from the enterprise's Super Radar transaction information.

[0063] When extracting characteristic data from the enterprise's qualification information, not only the enterprise's professional qualification certificate and other characteristic data are extracted from the enterprise's governing information, but also the various honors the enterprise has received, government funding support, financing support, tax incentives, project support, green production and certification and other characteristic data are extracted from the enterprise's government reward and subsidy information. In addition, characteristic data such as whether the enterprise participates in the formulation of industry standards can be extracted from the enterprise's standard-setting information.

[0064] When extracting characteristic data from the R&D information of an enterprise, characteristic data is extracted not only from the R&D funding information and R&D talent information of the enterprise, but also from the enterprise's patent characteristic data and software copyright characteristic data from the enterprise's intellectual property information.

[0065] When extracting characteristic data from the business information of an enterprise, multiple types of characteristic data such as the entity type, business scope, residence or main business location of the enterprise are extracted.

[0066] It can be seen that the manufacturing enterprise scoring method provided in this embodiment not only scores manufacturing enterprises based on multiple types of enterprise information of manufacturing enterprises (including financial information, credit information, business information, qualification information, R&D information and industrial and commercial information), but also has multiple scoring index dimensions. Moreover, each type of enterprise information includes at least two or more sub-types of enterprise information. When extracting feature data of each type of enterprise information, feature data of each sub-type of enterprise information in each type of enterprise information is extracted. In other words, the feature data extracted from each type of enterprise information shows the situation of the enterprise in this aspect as comprehensively as possible, so that the final scoring of the enterprise is more objective and accurate.

[0067] Optionally, in some embodiments of the present application, the enterprise scoring model includes an input layer, at least one hidden layer, and an output layer;

[0068] The input layer is used to perform weighted calculation on the feature data of each sub-type of enterprise information in each type of input enterprise information, wherein the weight of the feature data of each sub-type of enterprise information is a learnable network parameter and is fixed in the model training stage;

[0069] At least one hidden layer is used to perform feature processing on the output result of the input layer, and send the features of the last hidden layer output to the output layer;

[0070] The output layer is used to predict the ratings of manufacturing companies based on the input features.

[0071] Taking the R&D information of an enterprise as an example, the characteristic data such as the R&D funds, R&D talents, invention patents, utility model patents, design patents and software copyrights of the enterprise are extracted from the sub-type enterprise information such as the R&D funds, R&D talents and intellectual property information of the enterprise, which are expressed as {X1, X2, ···, X k}, the input layer performs weighted calculation on the characteristic data of each sub-type enterprise information in the input enterprise R&D information, which can be expressed as Z = A1×X1+ A2×X2+···+ A k ×X k , where {A1, A2, ···, A n} is {X1, X2, ···, X k The weights of the feature data in each sub-type of enterprise information are learnable network parameters and are fixed during the model training phase. That is to say, after the enterprise scoring model is trained, the input layer performs weighted calculations on the feature data of each sub-type of enterprise information in each type of input enterprise information, and the weights of the feature data of each sub-type of enterprise information are fixed.

[0072] It is understandable that after training the enterprise scoring model, the enterprise scoring model can be verified. Specifically, the characteristic data of various types of enterprise information (including capital information, credit information, business information, qualification information, R&D information, and industrial and commercial information) of other sample manufacturing enterprises are used as verification samples, and the scores of the other sample manufacturing enterprises are used as sample labels to verify the enterprise scoring model. If the verification result meets the preset verification conditions, the trained model is used as the enterprise scoring model. If the verification result does not meet the preset verification conditions, the model continues to be trained until the verification result meets the preset verification conditions. The verification condition can be at least one of accuracy, recall rate, and AUC.

[0073] Based on any of the above embodiments, optionally, in some embodiments of the present application, there are multiple manufacturing enterprises, and the aforementioned steps of the manufacturing enterprise scoring method can be used to score multiple manufacturing enterprises. In this embodiment, the manufacturing enterprise scoring method can also include:

[0074] S40: After obtaining the scores of each manufacturing enterprise, rank each manufacturing enterprise or divide it into different levels according to the scores, for example, divide it into ten levels.

[0075] In this way, advanced manufacturing enterprises can be easily selected, further facilitating account managers to create exclusive financial service solutions for them.

[0076] If there are multiple manufacturing enterprises, and the aforementioned steps of the manufacturing enterprise scoring method are used to score the multiple manufacturing enterprises, optionally, in some other embodiments of the present application, the manufacturing enterprise scoring method may further include:

[0077] S50: After obtaining the scores of each manufacturing enterprise, select the top n manufacturing enterprises with the highest scores in each manufacturing industry as the leading enterprises of the manufacturing industry for display, n≥1, and n is an integer.

[0078] In this way, the top n manufacturing companies with the highest scores in each manufacturing industry can be obtained, and a cluster map of leading companies in each manufacturing industry can be displayed, which makes it convenient for customer managers to obtain marketing guidance from the cluster map of leading companies in each manufacturing industry.

[0079] If there are multiple manufacturing enterprises, and the aforementioned steps of the manufacturing enterprise scoring method are used to score the multiple manufacturing enterprises, optionally, in some other embodiments of the present application, the manufacturing enterprise scoring method may further include:

[0080] S60: After obtaining the scores of each manufacturing enterprise, the top m manufacturing enterprises with the highest scores in each industrial park are selected as the key enterprises of the industrial park for display, where m≥1 and m is an integer.

[0081] In this way, the top m manufacturing enterprises with the highest scores in each industrial park can be obtained, and the key enterprises in each industrial park can be obtained, making it easier for account managers to obtain marketing guidance.

[0082] If there are multiple manufacturing enterprises, and the aforementioned steps of the manufacturing enterprise scoring method are used to score the multiple manufacturing enterprises, optionally, in some further embodiments of the present application, the manufacturing enterprise scoring method may further include:

[0083] S70: After obtaining the scores of each manufacturing enterprise, select the top x manufacturing enterprises with the highest scores in each major project as the leading enterprises of the major project for display, where x≥1 and x is an integer.

[0084] In this way, the top m manufacturing companies with the highest scores in each major project can be obtained, and the leading companies in each major project can be obtained, making it easier for account managers to obtain marketing guidance.

[0085] Accordingly, the present application also provides a manufacturing enterprise scoring device. Figure 2The structural block diagram of the manufacturing enterprise scoring device provided in the embodiment of the present application is shown as follows: Figure 2 As shown, the manufacturing enterprise scoring device includes:

[0086] The enterprise information acquisition unit 10 is used to acquire various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information;

[0087] A feature data extraction unit 20, used to extract feature data of each type of enterprise information;

[0088] The scoring unit 30 is used to input the characteristic data of each type of enterprise information into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels.

[0089] Each type of enterprise information may include at least two or more sub-types of enterprise information; when the feature data extraction unit 20 extracts the feature data of each type of enterprise information, it specifically extracts the feature data of each sub-type of enterprise information in each type of enterprise information.

[0090] Since the manufacturing enterprise scoring device and the above-mentioned manufacturing enterprise scoring method correspond to each other, and the various steps of the manufacturing enterprise scoring method provided in the embodiment of the present application have been discussed in detail above, they will not be repeated here.

[0091] The embodiment of the present application also provides an electronic device, Figure 3 The hardware structure block diagram of the electronic device provided in the embodiment of the present application is shown as follows: Figure 3 As shown, the electronic device includes at least one processor and a memory connected to the processor, and may include: at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4;

[0092] In the embodiment of the present application, the number of the processor 1, the communication interface 2, the memory 3, and the communication bus 4 is at least one, and the processor 1, the communication interface 2, and the memory 3 communicate with each other through the communication bus 4;

[0093] The processor 1 may be any one or more of a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP) or a digital signal processor (DSP);

[0094] The memory 3 may include a volatile memory (volatile memory), such as a random access memory (RAM); and may also include a non-volatile memory (non-volatile memory), such as a read-only memory (ROM), a flash memory, a hard drive (HDD) or a solid state drive (SSD).

[0095] The memory 3 is used to store a computer program, and the processor 1 is used to execute the computer program so that the electronic device can achieve:

[0096] Obtaining various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information;

[0097] Extract characteristic data of each type of enterprise information;

[0098] The characteristic data of each type of enterprise information is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of sample manufacturing enterprises as sample labels.

[0099] Optionally, the detailed functions and extended functions of the computer program may refer to the above description and will not be repeated here.

[0100] The present application also provides a computer storage medium, wherein the storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement:

[0101] Obtaining various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information;

[0102] Extract characteristic data of each type of enterprise information;

[0103] The characteristic data of each type of enterprise information is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of sample manufacturing enterprises as sample labels.

[0104] Optionally, the detailed functions and extended functions of the computer program may refer to the above description and will not be repeated here.

[0105] In summary, the manufacturing enterprise scoring method and related devices provided in the embodiments of the present application first obtain various types of enterprise information of the manufacturing enterprise to be scored, wherein the various types of enterprise information include financial information, credit information, operating information, qualification information, research and development information, and industrial and commercial information; then extract the characteristic data of each type of enterprise information; finally, input the characteristic data of each type of enterprise information into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model, so as to facilitate the selection of advanced manufacturing enterprises and create financial service solutions for advanced manufacturing enterprises; wherein the enterprise scoring model is trained using the characteristic data of various types of enterprise information of the sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels. The manufacturing enterprise scoring method and related devices provided in the embodiments of the present application score manufacturing enterprises based on multiple types of enterprise information of manufacturing enterprises (including financial information, credit information, operating information, qualification information, research and development information, and industrial and commercial information). The scoring indicators have multiple dimensions and a wide range of data sources, so that the scoring of manufacturing enterprises is more comprehensive, objective, and accurate. In addition, the feature data extracted from each type of enterprise information of the manufacturing enterprises is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the feature data of each type of enterprise information of the sample manufacturing enterprises as training samples and the scores of the sample manufacturing enterprises as sample labels. That is, the enterprise scoring model is determined by training a machine learning model based on big data, which is conducive to improving the accuracy and credibility of the scoring of manufacturing enterprises.

[0106] The various parts in this manual are described in a combination of parallel and progressive ways. Each part focuses on the differences from other parts, and the same or similar parts between the various parts can be referenced to each other.

[0107] With respect to the above description of the disclosed embodiments, the features described in the embodiments in this specification may be replaced or combined with each other, so that professionals in the field can implement or use the present application. Various modifications to these embodiments will be apparent to professionals in the field, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A manufacturing enterprise scoring method, characterized in that: include: Obtaining various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information; Extract characteristic data of each type of enterprise information; The characteristic data of each type of enterprise information is input into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of sample manufacturing enterprises as sample labels.

2. The manufacturing enterprise scoring method according to claim 1 is characterized in that: Each type of enterprise information includes at least two or more sub-types of enterprise information; The characteristic data extracted for each type of enterprise information include: Extract characteristic data of each sub-type of enterprise information in each type of enterprise information.

3. The manufacturing enterprise scoring method according to claim 2 is characterized in that: The enterprise scoring model includes an input layer, at least one hidden layer and an output layer; The input layer is used to perform weighted calculation on the feature data of each sub-type of enterprise information in each type of input enterprise information, wherein the weight of the feature data of each sub-type of enterprise information is a learnable network parameter and is fixed in the model training stage; The at least one hidden layer is used to perform feature processing on the result output by the input layer, and send the feature output by the last hidden layer to the output layer; The output layer is used to predict the rating of the manufacturing enterprise based on the input features.

4. The manufacturing enterprise scoring method according to any one of claims 1 to 3, characterized in that: There are multiple manufacturing enterprises; the manufacturing enterprise scoring method also includes: After obtaining the scores of each of the manufacturing enterprises, the manufacturing enterprises are ranked or divided into different levels according to the scores.

5. The manufacturing enterprise scoring method according to any one of claims 1 to 3, characterized in that: There are multiple manufacturing enterprises; the manufacturing enterprise scoring method also includes: After obtaining the scores of each of the manufacturing enterprises, the top n manufacturing enterprises with the highest scores in each manufacturing industry are selected as the leading enterprises of the manufacturing industry for display, where n≥1 and n is an integer.

6. The manufacturing enterprise scoring method according to any one of claims 1 to 3, characterized in that: There are multiple manufacturing enterprises; the manufacturing enterprise scoring method also includes: After obtaining the scores of the manufacturing enterprises, the top m manufacturing enterprises with the highest scores in each industrial park are selected as the key enterprises of the industrial park for display, where m≥1 and m is an integer.

7. The manufacturing enterprise scoring method according to any one of claims 1 to 3, characterized in that: There are multiple manufacturing enterprises; the manufacturing enterprise scoring method also includes: After obtaining the scores of each of the manufacturing enterprises, the top x manufacturing enterprises with the highest scores in each major project are selected as the leading enterprises of the major project for display, where x≥1 and x is an integer.

8. A manufacturing enterprise scoring device, characterized in that: include: An enterprise information acquisition unit is used to acquire various types of enterprise information of the manufacturing enterprise to be rated, wherein the various types of enterprise information include capital information, credit information, operation information, qualification information, research and development information, and industrial and commercial information; A feature data extraction unit, used to extract feature data of each type of enterprise information; The scoring unit is used to input the characteristic data of each type of enterprise information into the configured enterprise scoring model to obtain the score of the manufacturing enterprise output by the enterprise scoring model. The enterprise scoring model is trained using the characteristic data of each type of enterprise information of sample manufacturing enterprises as training samples and the scores of sample manufacturing enterprises as sample labels.

9. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the manufacturing enterprise scoring method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the manufacturing enterprise scoring method as described in any one of claims 1 to 7.

Images