Enterprise credit granting processing method, device, equipment, medium and product

Abstract

The invention discloses an enterprise credit granting processing method, device and equipment, a medium and a product, and relates to the field of financial science and technology. The method comprises the following steps: acquiring public financial data and public property data of a target enterprise and industry technical information of a target industry; determining at least one enterprise technology point of the target enterprise according to the public known production data and the industry technology information; determining the attention weight of each enterprise technology point relative to each technology hotspot in the industry technology information based on the public known production data and each enterprise technology point; and determining a credit granting result of the target enterprise according to the public financial data, the technical point of each enterprise and the attention weight. According to the technical scheme of the embodiment of the invention, the multi-dimensional heterogeneous data processing efficiency and the credit extension accuracy in the credit extension process can be effectively improved, and the method has good universality and adaptability.

Description

Technical Field

[0001] This application relates to the field of financial technology, and in particular to a method, apparatus, equipment, medium and product for corporate credit processing. Background Technology

[0002] With societal progress and significant advancements in science and technology, an increasing number of industries are adopting emerging technologies in data processing. In enterprise technology assessment and analysis, traditional methods often face challenges in data integration and quantitative correlation. Existing methods typically examine an enterprise's own technological assets and the overall technological dynamics of the industry separately, but these two aspects are often fragmented, making it difficult to systematically measure their inherent connections.

[0003] Currently, effectively linking and analyzing scattered, multi-source, multi-type, and multi-constructed data to extract representative quantitative indicators is a common technical challenge. Furthermore, the integration and calculation of these indicators with data from other dimensions lacks flexible and adaptive mechanisms. Therefore, constructing an analytical framework capable of automating multi-source data processing, accurately quantifying correlations, and supporting dynamic weight adjustments has become a key research focus for professionals in related fields. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, medium, and product for enterprise credit processing, in order to improve the accuracy, objectivity, and rationality of credit processing.

[0005] According to one aspect of this application, a corporate credit processing method is provided, comprising: Obtain publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry; Based on publicly available intellectual property data and industry technical information, identify at least one technical point of the target company. Based on publicly available intellectual property data and the technical points of each enterprise, the attention weight of each enterprise's technical points relative to the various technical hotspots in industry technical information is determined. The credit granting results for the target companies are determined based on publicly available financial data, the technical strengths of each company, and the attention weighting.

[0006] According to another aspect of this application, a corporate credit processing apparatus is provided, comprising: The data acquisition module is used to acquire publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry. The Enterprise Technology Point Determination Module is used to determine at least one enterprise technology point of a target enterprise based on publicly available intellectual property data and industry technology information. The attention weight determination module is used to determine the attention weight of each company's technology points relative to the various technology hotspots in the industry's technical information, based on publicly available intellectual property data and the technology points of each company. The credit result determination module is used to determine the credit result of the target company based on publicly available financial data, the technical points of each company, and attention weights.

[0007] According to another aspect of this application, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the enterprise credit processing method described in any embodiment of this application.

[0008] According to another aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the enterprise credit processing method described in any embodiment of this application.

[0009] According to another aspect of this application, a computer program product is provided, the computer program product including a computer program that, when executed by a processor, implements the enterprise credit processing method according to any embodiment of this application.

[0010] In the technical solution of this application embodiment, at least one key technical point of the target enterprise is extracted based on its publicly available intellectual property data and industry technical information. This solves the technical challenge of automatically identifying and structuring enterprise technical characteristics from multi-source heterogeneous data, improving the efficiency of processing multi-source heterogeneous data during credit granting, and realizing a quantitative representation of the enterprise's technical positioning within the industry, providing input based on objective technical facts for subsequent analysis. By introducing an attention mechanism, the weight of each enterprise's technical point relative to industry technical hotspots is calculated, solving the technical problems of difficulty in quantifying technical relevance and the inability to dynamically adjust importance in traditional evaluation methods, thereby improving the distinguishability and interpretability of the technical dimension in credit granting. Combining publicly available financial data, enterprise technical points, and their attention weights, a credit granting mechanism is constructed, effectively overcoming the limitations of single financial indicator evaluation and improving the accuracy, objectivity, and robustness of credit granting. Through an end-to-end computable framework, this solution ensures that the mechanism remains stable and usable in the face of data updates and dynamic industry changes, exhibiting strong generalization ability and adaptability.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a corporate credit processing method provided in Embodiment 1 of this application; Figure 2 This is a flowchart of a corporate credit processing method provided according to Embodiment 2 of this application; Figure 3 This is a schematic diagram of the structure of an enterprise credit processing device according to Embodiment 3 of this application; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the enterprise credit processing method of the embodiments of this application; Detailed Implementation

[0014] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] Example 1 Figure 1This application provides a flowchart of a corporate credit processing method according to Embodiment 1. This embodiment is applicable to processing publicly available financial data and publicly available intellectual property data of a company and determining the credit granting result. This method can be executed by a corporate credit processing device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes: S110. Obtain publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry.

[0017] The target company can be any enterprise that requires credit approval. Therefore, data corresponding to the target company is processed to determine the credit approval result. Publicly available financial data can be financial data disclosed by the target company to the public, such as legally disclosed financial statements and audit reports. This financial data is information publicly available to the public. Publicly available intellectual property data can be the target company's intellectual property data, such as patent application and authorization data, copyright data, and trademark data, etc., which are not exhaustively listed here. It is understood that due to the nature of intellectual property, the intellectual property data of enterprises are all publicly available. Therefore, the publicly available financial data and publicly available intellectual property data of the target company can be directly obtained through the Internet and other means. Industry technical information can be information on the current state of technology in the industry in which the target company operates, such as information on the current state of technology and technological hotspots in the industry. Industry technical information can be identified and obtained based on publicly available industry standard documents or industry-related web pages and news content on the Internet. This application embodiment does not limit the content and acquisition method of industry technical information. The description of this application embodiment and its various implementation methods mainly focuses on how to process and analyze publicly available financial data, publicly available intellectual property data, and industry technical information. It is understandable that publicly available financial data, publicly available intellectual property data, and different industry technical information are multi-dimensional, multi-source, and heterogeneous data. They may be structured or unstructured data, and the carriers (text, images, etc.) on which the data is carried may also be different. Therefore, the technical solution of the embodiments of this application can also effectively improve the processing efficiency of these multi-source data in the credit granting process.

[0018] S120. Based on publicly available intellectual property data and industry technical information, identify at least one technical point of the target company.

[0019] Among these, enterprise technology points can refer to the technical information of various related technologies that the target enterprise uses and is involved in in its participation in social production. It is understandable that publicly available intellectual property data (especially patent information) reflects a large amount of technology developed and used by the target enterprise. On the other hand, industry technology information of the target enterprise's industry includes all the technology points that may be involved in that industry. Therefore, the various enterprise technology points involved by the target enterprise can be analyzed from its publicly available intellectual property data and industry technology information.

[0020] For example, structured data can be extracted from unstructured data such as publicly available intellectual property data and industry technical information using techniques such as large language models or semantic analysis. This allows for the extraction of necessary technical keywords, which can then be used to determine which technologies the target company employs. Alternatively, semantic recognition can be performed on the text in publicly available intellectual property data and industry technical information, and the results can be converted into semantic vectors. Further analysis of the semantic similarity between these vectors can then determine which technologies within the target company's industry it employs, thus identifying these technologies as the target company's key technical features.

[0021] S130. Based on publicly available intellectual property data and the technical points of each enterprise, determine the attention weight of each enterprise's technical points relative to the technical hotspots in the industry's technical information.

[0022] Among them, "technological hotspots" can refer to information on technological directions within the industry that have attracted significant user attention, widespread discussion, substantial resource investment, or are being promoted and implemented. Technological hotspots, to a certain extent, represent the trends and directions of technological development and promotion within the industry. Attention weight can be calculated using an attention mechanism to assign weights to a target company's technological points relative to various technological hotspots within the industry; that is, the importance or relevance of the company's technological points to the technological hotspots. It can be understood that the higher the attention weight of a company's technological point, the more aligned that point is with the direction of industry technological development. In other words, the higher the attention weight, the more well-suited and relevant the company's technological point is to the development of industry technological hotspots, indirectly indicating that the company's technological point is more important to the target company. Of course, the attention mechanism algorithm used in this embodiment is not limited here.

[0023] S140. Determine the credit granting results for the target company based on publicly available financial data, the technical points of each company, and the attention weights.

[0024] Understandably, publicly available financial data can reveal the financial status of each technology point of the target company, such as the target company's input and output for that technology point. Credit approval can be the result of a credit endorsement granted to the target company, which can be used for the company's work within the industry (e.g., participating in major projects, bidding, etc.) and for loans.

[0025] The proportion of each company's technological features reflected in its publicly available financial data can characterize the contribution of these technologies to the target company. Meanwhile, attention weights can characterize the degree of alignment between these technologies and industry development. Based on these two aspects, the quality of the target company's financial performance in relation to industry hot technologies can be further determined, thereby enabling the assessment of the target company's creditworthiness.

[0026] In the technical solution of this application embodiment, at least one key technical point of the target enterprise is extracted based on its publicly available intellectual property data and industry technical information. This solves the technical challenge of automatically identifying and structuring enterprise technical characteristics from multi-source heterogeneous data, improving the efficiency of processing multi-source heterogeneous data during credit granting, and realizing a quantitative representation of the enterprise's technical positioning within the industry, providing input based on objective technical facts for subsequent analysis. By introducing an attention mechanism, the weight of each enterprise's technical point relative to industry technical hotspots is calculated, solving the technical problems of difficulty in quantifying technical relevance and the inability to dynamically adjust importance in traditional evaluation methods, thereby improving the distinguishability and interpretability of the technical dimension in credit granting. Combining publicly available financial data, enterprise technical points, and their attention weights, a credit granting mechanism is constructed, effectively overcoming the limitations of single financial indicator evaluation and improving the accuracy, objectivity, and robustness of credit granting. Through an end-to-end computable framework, this solution ensures that the mechanism remains stable and usable in the face of data updates and dynamic industry changes, exhibiting strong generalization ability and adaptability.

[0027] Example 2 Figure 2 This is a flowchart illustrating a corporate credit processing method provided in Embodiment 2 of this application. This embodiment further refines the determination of attention weights based on the aforementioned embodiments. Figure 2 As shown, the method includes: S210. Obtain publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry.

[0028] S220. Based on publicly available intellectual property data and industry technical information, identify at least one enterprise technology point of the target enterprise.

[0029] S230. Determine the feature vector of each enterprise's technology point and the feature vector of each technology hotspot.

[0030] The enterprise technology point feature vector can be the feature vector corresponding to the information of the enterprise technology point, and similarly, the technology hotspot feature vector can be the feature vector corresponding to the information of the technology hotspot. For example, the information of enterprise technology points and technology hotspots can be converted into embedding vectors, and the information of enterprise technology points and technology hotspots can be semantically embedded by a trained language processing model to obtain the corresponding two feature vectors. Of course, the embodiments of this application do not limit the language processing model (especially natural language processing model) for implementing the semantic embedding function.

[0031] S240. Based on publicly available intellectual property data and the technical points of each enterprise, determine the external weight of the attention of each enterprise's technical points relative to each technical hotspot.

[0032] External weights can be additional weighting factors used to provide external support for feature vectors in the attention mechanism. The attention weights in the attention mechanism can be obtained by fusing the original attention score and the external weights, and the fusing method can be normalization. In this embodiment, the purpose of the attention mechanism is to explore the correlation between each company's technical points and various technological hotspots, that is, which company's technical points are more closely aligned with industry technological hotspots. The setting of external weights can emphasize the importance of different company technical points based on real-world conditions, which can be determined based on publicly available intellectual property data and the technical points of each company.

[0033] The external weight of each enterprise's technology point can be the degree of importance of that technology point to the enterprise. It is understood that the publicly available intellectual property data of an enterprise corresponds to different enterprise technologies of the target enterprise. The quantity and quality of the intellectual property data corresponding to different enterprise technologies determine the importance of these different enterprise technologies to the enterprise. Therefore, by analyzing the publicly available intellectual property data of the target enterprise, the contribution of this intellectual property data to the enterprise's technology points is determined, and this is used to calculate the external weight. Of course, this contribution can be quantified based on the publicly available intellectual property data, such as the number of patents and patent strength corresponding to each enterprise technology point. The number of patents can be obtained by filtering and statistically analyzing the publicly available intellectual property data, and the patent strength can be obtained from publicly available databases or websites storing patent information. For example, the number of patents or the patent strength can be normalized, and the normalization result serves as the external weight of each enterprise technology point. Of course, this normalization process can employ any normalization algorithm from related technologies, and this application embodiment does not limit this.

[0034] S250. Determine the attention weight based on the enterprise's technical point feature vector, technical hotspot feature vector, and external weights.

[0035] It is understood that the feature vectors of enterprise technology points and feature vectors of technology hotspots are processed into vectors, and based on a preset attention mechanism, the external weights determined in the aforementioned steps are incorporated for calculation to obtain the attention weight of each enterprise's technology point relative to the industry's technology hotspots. This application embodiment does not limit the specific form and algorithm of this attention mechanism.

[0036] S260. Determine the credit granting results for the target company based on publicly available financial data, the technical strengths of each company, and the attention weights.

[0037] As described in the preceding steps, attention weights characterize the relevance of each enterprise's technological points to industry trends. Publicly available financial data can differentiate the financial situations corresponding to different enterprise technological points across various business lines or product lines. Combining these two pieces of information allows for targeted analysis of the financial situation of the target enterprise's use of industry-leading technologies. This enables the assessment of the target enterprise's prospects within the industry, thereby obtaining industry credit endorsement. Of course, the specific credit granting result can be diverse, such as a credit line. Publicly available financial data, information on each enterprise's technological points, and the attention weights determined in the preceding steps can be input into a pre-trained credit result evaluation model. This model outputs the target enterprise's credit granting result. This credit result evaluation model can employ any machine learning model from relevant technologies and be pre-trained using historical enterprise data and historical credit granting results; this application embodiment does not limit this approach.

[0038] In the technical solution of this application embodiment, by successively determining the enterprise's technical point feature vector, the technical hotspot feature vector, and the external weight of the enterprise's technical points, and further determining the credit granting result, this process not only vectorizes the information of the enterprise's technical points and the industry's technical hotspots and specifically participates in the data processing process, but also determines the external weight of the enterprise's technical points and introduces an attention mechanism, enabling quantifiable data processing between the enterprise's technical points and the industry's technical hotspots, making the process of determining the credit granting result more objective and reliable.

[0039] In an optional implementation, determining the attention weight based on the enterprise's technology point feature vector, technology hotspot feature vector, and external weights in step S250 may include: S251. Determine the original attention score based on the enterprise's technical point feature vector and technical hotspot feature vector.

[0040] The original attention score can be an intermediate value in the attention mechanism, representing the unnormalized correlation strength between the query vector and the key vector. In the example of this application embodiment, the enterprise technology point feature vector can be used as the query vector in the attention mechanism, and correspondingly, the technology hotspot vector can be used as the key vector. Generally, the original attention score can be calculated using dot product and scaling. For a given query vector and key vector, the similarity between the two vectors is measured by calculating their dot product, and then scaling is used to prevent the dot product result from becoming too large and causing gradient vanishing. The original attention score in the attention mechanism calculation process is obtained through dot product and scaling. The scaling scale can be determined according to the dimension of the key vector, which is not limited in this application embodiment.

[0041] S252. Determine the attention weights based on the original attention scores and external weights.

[0042] In the attention mechanism, the attention weight of the enterprise's technical points relative to the industry's technical hotspots is calculated based on the original attention score and external weights determined in the aforementioned steps. The original attention score and external weights are first fused, and then normalized to obtain the attention weights. The fusion method can be additive fusion, multiplicative fusion, gating fusion, etc., and this application embodiment does not limit the fusion method and normalization method.

[0043] In the above implementation, the initial attention between the enterprise's technical point feature vector and the technical hotspot feature vector is first determined. Then, combined with the external weights determined in the aforementioned implementation, the attention weight is calculated to provide a specific basis for subsequent credit granting processing. By converting the information of the enterprise's technical points and technical hotspots into feature vectors for processing, and integrating and processing different multi-source heterogeneous data, the processing efficiency of multi-source heterogeneous data in the credit granting process can be improved. It can be understood that this attention weight is calculated based on the target enterprise's technical points and the industry hotspots of its sector. These bases are substantial, thus helping to improve the objectivity and rationality of credit granting processing.

[0044] In an optional implementation, the determination of the external weight of attention for each enterprise's technology point relative to each technology hotspot based on publicly available intellectual property data and each enterprise's technology points in step S240 may include: S241. Obtain the R&D investment ratio of each technology point in the target company.

[0045] The R&D investment ratio data can be the proportion of the target company's R&D investment in each technology area out of its total R&D investment. This data can be obtained by statistical analysis and calculation from the company's publicly available financial statements or R&D data. Understandably, the more a target company invests in a particular technology area, the more it indicates the degree of understanding the target company has of that technology, and also briefly illustrates the importance of that technology to the target company.

[0046] S242. Determine the amount of intellectual property authorization corresponding to each enterprise's technical points in the publicly available intellectual property data.

[0047] Among these, intellectual property licensing can refer to the number of patents granted to a target company for its own intellectual property, such as the number of patents granted, the number of software copyrights granted, and the number of trademarks granted. It is understandable that the number of intellectual property licenses corresponding to different companies' technological points varies, and the number of intellectual property licenses corresponding to different companies' technological points can, to some extent, reflect the target company's research inclinations and capabilities in different technological areas.

[0048] S243. For any enterprise technology point, determine the support strength value of the enterprise technology point based on the R&D investment ratio data and the amount of intellectual property authorization.

[0049] Generally, in attention mechanisms, the support strength value specifically refers to a quantitative indicator retrieved or calculated from external knowledge sources to measure the reliability of a piece of information in supporting the current context. In the embodiments and implementations of this application, the R&D investment ratio and the amount of intellectual property licenses are the external knowledge sources for the attention mechanism. From the R&D investment ratio and the amount of intellectual property licenses, we can determine the importance of each enterprise's technology points to the target enterprise, that is, determine the support strength value of each enterprise's technology points. The support strength value can be quantitatively calculated directly by normalizing the values ​​of the R&D investment ratio and the amount of intellectual property licenses. Of course, the embodiments of this application do not limit the method and specific algorithm of this normalization process.

[0050] S244. Use the vector formed by the support strength values ​​of all enterprise technical points as the external weight.

[0051] The support strength values ​​corresponding to the different enterprise technology points determined in the aforementioned steps are combined into a vector form and used as external weights to participate in the calculation of the attention mechanism.

[0052] In the above implementation, by processing the R&D investment ratio data and the amount of intellectual property authorization, the support strength value for generating external weights is determined. This support strength value effectively characterizes the importance of different enterprise technology points to the target enterprise, providing a basis for the external weights of different enterprise technology points in the subsequent calculation of attention weights, and improving the objectivity and accuracy of attention weight calculation.

[0053] In an optional implementation, step S260, determining the creditworthiness of the target company based on publicly available financial data, the technical strengths of each company, and attention weights, may include: S261. Extract the business line financial data corresponding to the technical points of each enterprise from the publicly available financial data.

[0054] The business line financial data can be the financial data corresponding to different business lines of the target company. It is understood that a company may have multiple different business lines (e.g., different product lines, different technical service lines, etc.), and different business lines employ different technologies. The goal is to identify which financial data belongs to which company's technology point and corresponding business line from publicly available financial data, thereby distinguishing the financial data of different business lines. For example, natural language processing algorithms can be used to identify the content in publicly available financial data, determine its correlation with each company's technology point or different business lines, and thus distinguish the financial data of different business lines. Of course, any natural language processing algorithm from related technologies can be used for processing; this application embodiment does not limit this.

[0055] S262. Based on a pre-trained semantic embedding model, convert the financial data of each business line and the technical points of each enterprise into business line financial vectors and enterprise technical point vectors, and determine the cosine similarity between the business line financial vectors and enterprise technical point vectors.

[0056] Semantic embedding models are artificial intelligence models that convert text (words, sentences, paragraphs, or even documents) into numerical vectors. These vectors are characterized by semantically similar text, meaning their corresponding vectors are closer in mathematical space. Cosine similarity is an indicator used to measure the directional similarity between two vectors. Therefore, a pre-trained semantic embedding model is used to vectorize business line financial data and enterprise technical information, resulting in business line financial vectors and enterprise technical vectors. Then, the directional similarity between these two vectors in mathematical space is calculated, and finally, the cosine similarity between the two vectors in mathematical space is calculated.

[0057] S263. Determine the credit granting result for the target enterprise based on cosine similarity and attention weight.

[0058] The target company's credit rating, serving as a quantitative indicator of corporate credit endorsement, can be calculated using the cosine similarity and attention weight determined in the preceding steps. For example, the values ​​of cosine similarity and attention weight can be used as two types of coefficients, multiplied by a pre-set baseline credit rating to obtain the target company's credit rating. It can be understood that a higher cosine similarity between the business line financial vector and the company's technology vector indicates that the technology the target company focuses on is precisely the technology upon which its financially strong business lines rely; a higher attention weight for the company's technology points relative to industry technology hotspots indicates that the target company's technology layout is more aligned with industry development trends, possessing better matching and foresight. Based on the pre-set baseline credit rating, the quantitative value of the target company's credit endorsement is calculated by influencing it with the two types of coefficients: cosine similarity and attention weight.

[0059] In the above implementation, business line financial data is extracted from publicly available financial data, and the business line financial data and enterprise technical information are vectorized respectively. The cosine similarity between the two is then calculated to assess their correlation. Vectorizing different textual and numerical information to explore the correlation between business line financial data and enterprise technical points can effectively improve the efficiency of processing multi-source heterogeneous data in the credit granting process, and obtain the basis that can support the determination of credit granting results, which helps to improve the accuracy of credit granting results.

[0060] In one optional implementation, the step of determining at least one enterprise technology point of the target enterprise based on publicly available intellectual property data and industry technical information in S220 may include: S221. Based on the trained text vector model, publicly available intellectual property data and industry technical information are transformed into intellectual property text vectors and industry information vectors, respectively.

[0061] The text vector model can be any model for vectorizing text. It vectorizes the text in publicly available intellectual property data and the text in industry technical information to obtain intellectual property text vectors and industry information vectors, respectively. This application's implementation does not limit the text vectorization process.

[0062] S222. Calculate the semantic similarity between the intellectual property text vector and the industry information vector.

[0063] Based on a pre-defined semantic similarity algorithm, the semantic similarity between intellectual property text vectors and industry information vectors is calculated. Through the calculation of semantic similarity, the similarity between the two types of texts is determined.

[0064] S223. In response to the semantic similarity exceeding the preset similarity threshold, determine the enterprise technology points of the target enterprise.

[0065] It is understandable that when the semantic similarity between the intellectual property text vector and the industry information vector reaches a certain level, it is possible to determine the text content that matches the publicly available intellectual property data and industry technical information of the target company, thereby further determining which technical points in the industry the target company has adopted, and taking these technical points as the target company's corporate technical points.

[0066] In the above implementation, the text in the publicly available intellectual property data and industry technical information is vectorized and semantic similarity is calculated based on the vectors. The technical content in the publicly available intellectual property data that is similar to the industry technical information is identified as the enterprise technical points of the target enterprise. Through vectorized similarity processing, the difficulty of identifying similar information in publicly available intellectual property data and industry technical information can be simplified, and the efficiency and accuracy of enterprise technical point identification can be improved.

[0067] Example 3 Figure 3 This is a schematic diagram of a corporate credit processing device provided in Embodiment 3 of this application. Figure 3 As shown, the device includes: The data acquisition module 310 is used to acquire publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry. The Enterprise Technology Point Determination Module 320 is used to determine at least one enterprise technology point of the target enterprise based on publicly available intellectual property data and industry technology information. The attention weight determination module 330 is used to determine the attention weight of each enterprise's technology points relative to the various technology hotspots in the industry's technical information, based on publicly available intellectual property data and the technology points of each enterprise. The credit result determination module 340 is used to determine the credit result of the target company based on publicly available financial data, the technical points of each company, and attention weights.

[0068] In the technical solution of this application embodiment, at least one key technical point of the target enterprise is extracted based on its publicly available intellectual property data and industry technical information. This solves the technical challenge of automatically identifying and structuring enterprise technical characteristics from multi-source heterogeneous data, improving the efficiency of processing multi-source heterogeneous data during credit granting, and realizing a quantitative representation of the enterprise's technical positioning within the industry, providing input based on objective technical facts for subsequent analysis. By introducing an attention mechanism, the weight of each enterprise's technical point relative to industry technical hotspots is calculated, solving the technical problems of difficulty in quantifying technical relevance and the inability to dynamically adjust importance in traditional evaluation methods, thereby improving the distinguishability and interpretability of the technical dimension in credit granting. Combining publicly available financial data, enterprise technical points, and their attention weights, a credit granting mechanism is constructed, effectively overcoming the limitations of single financial indicator evaluation and improving the accuracy, objectivity, and robustness of credit granting. Through an end-to-end computable framework, this solution ensures that the mechanism remains stable and usable in the face of data updates and dynamic industry changes, exhibiting strong generalization ability and adaptability.

[0069] In one alternative implementation, the attention weight determination module 330 may include: The feature vector determination unit is used to determine the feature vector of each enterprise's technology point and the feature vector of each technology hotspot. The external weight determination unit is used to determine the external weight of each enterprise's technology points relative to each technology hotspot based on publicly available intellectual property data and each enterprise's technology points. The attention weight determination unit is used to determine attention weights based on the enterprise's technical point feature vector, technical hotspot feature vector, and external weights.

[0070] In one optional implementation, the attention weight determination unit may include: The original attention determination subunit is used to determine the original attention score based on the enterprise's technology point feature vector and technology hotspot feature vector; The attention weight determination subunit is used to determine the attention weights based on the original attention score and external weights.

[0071] In one optional implementation, the external weight determination unit may include: The percentage data determines the sub-unit, which is used to obtain the R&D investment percentage data of each enterprise's technology points in the target enterprise; The authorization quantity determination subunit is used to determine the intellectual property authorization quantity corresponding to each enterprise's technical point in the publicly available intellectual property data; The support strength determination subunit is used to determine the support strength value of any enterprise's technology point based on the R&D investment ratio data and the amount of intellectual property authorization. The external weight determination sub-unit is used to form a vector of support strength values ​​for all enterprise technical points, which serves as the external weight.

[0072] In one optional implementation, the credit granting result determination module 340 may include: The financial data extraction unit is used to extract business line financial data corresponding to the technical points of each enterprise from publicly available financial data. The similarity determination unit is used to convert the financial data of each business line and the technical points of each enterprise into business line financial vectors and enterprise technical point vectors based on the pre-trained semantic embedding model, and determine the cosine similarity between the business line financial vectors and enterprise technical point vectors. The credit result determination unit is used to determine the credit result of the target enterprise based on cosine similarity and attention weight.

[0073] In one optional implementation, the enterprise technology point determination module 320 may include: The vectorization unit is used to transform publicly available intellectual property data and industry technical information into intellectual property text vectors and industry information vectors, respectively, based on a text vector model trained with [training name missing]. The semantic similarity determination unit is used to calculate the semantic similarity between intellectual property text vectors and industry information vectors; The technology point determination unit is used to determine the enterprise technology points of the target enterprise in response to the semantic similarity exceeding a preset similarity threshold.

[0074] The enterprise credit processing device provided in this application embodiment can execute the enterprise credit processing method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing each enterprise credit processing method.

[0075] Example 4 Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of this application, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.

[0076] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory 12 or a random access memory 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 12 or loaded from storage unit 18 into the random access memory 13. The random access memory 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, read-only memory 12, and random access memory 13 are interconnected via a bus 14. An input / output interface 15 is also connected to the bus 14.

[0077] Multiple components in electronic device 10 are connected to input / output interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0078] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as enterprise credit processing methods.

[0079] In some embodiments, the enterprise credit processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory 12 and / or communication unit 19. When the computer program is loaded into random access memory 13 and executed by processor 11, one or more steps of the enterprise credit processing method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the enterprise credit processing method by any other suitable means (e.g., by means of firmware).

[0080] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), system-on-a-chip (SoCs), payload programmable logic devices (PLCs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0081] Computer programs used to implement the methods of this application may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0082] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0083] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a monitor with a cathode ray tube or liquid crystal display) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0084] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0085] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product within the cloud computing service system to address the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.

[0086] This application also discloses a computer program product, which includes a computer program that, when executed by a processor, implements the enterprise credit processing method provided in any embodiment of this application. This program product shares the same inventive concept as the enterprise credit processing methods disclosed in the embodiments of this application, and therefore will not be described in detail here.

[0087] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this application can be achieved, and this is not limited herein.

[0088] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for processing corporate credit, characterized in that, include: Obtain publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry; Based on the publicly available intellectual property data and the industry technical information, at least one enterprise technology point of the target enterprise is determined; Based on the publicly available intellectual property data and the technical points of each enterprise, the attention weight of each enterprise's technical point relative to the technical hotspots in the industry technical information is determined; The credit granting result for the target company is determined based on the publicly available financial data, the technical points of each company, and the attention weight.

2. The method according to claim 1, characterized in that, The step of determining the attention weight of each enterprise's technology point relative to the various technological hotspots in the industry technology information, based on the publicly available intellectual property data and the technology points of each enterprise, includes: Determine the feature vector of each enterprise technology point and the feature vector of each technology hotspot respectively; Based on the publicly available intellectual property data and the technical points of each enterprise, determine the external weight of attention for each technical point of the enterprise relative to each technical hotspot; The attention weight is determined based on the enterprise's technical feature vector, the technical hotspot feature vector, and the external weight.

3. The method according to claim 2, characterized in that, The step of determining the attention weight based on the enterprise's technical point feature vector, the technical hotspot feature vector, and the external weight includes: The original attention score is determined based on the enterprise's technical feature vector and the technical hotspot feature vector. The attention weight is determined based on the original attention score and the external weight.

4. The method according to claim 2, characterized in that, The determination of the external weight of attention for each of the enterprise's technological points relative to each of the technological hotspots, based on the publicly available intellectual property data and the technological points of each enterprise, includes: Obtain data on the R&D investment ratio of each technology point in the target company; Determine the amount of intellectual property authorization corresponding to each of the enterprise's technical points in the publicly available intellectual property data; For any enterprise technology point, the support strength value of the enterprise technology point is determined based on the R&D investment ratio data and the intellectual property authorization amount; The vector formed by the support strength values ​​of all enterprise technical points is used as the external weight.

5. The method according to claim 1, characterized in that, The process of determining the credit granting result for the target enterprise based on the publicly available financial data, the technical aspects of each enterprise, and the attention weight includes: Extract the business line financial data corresponding to each of the enterprise's technical points from the publicly available financial data; Based on a pre-trained semantic embedding model, the financial data of each business line and the technical points of each enterprise are converted into business line financial vectors and enterprise technical point vectors, and the cosine similarity between the business line financial vectors and the enterprise technical point vectors is determined. The credit granting result for the target enterprise is determined based on the cosine similarity and the attention weight.

6. The method according to any one of claims 1-5, characterized in that, The step of determining at least one enterprise technology point of the target enterprise based on the publicly available intellectual property data and the industry technology information includes: Based on the trained text vector model, the publicly available intellectual property data and the industry technical information are respectively transformed into intellectual property text vectors and industry information vectors. Calculate the semantic similarity between the intellectual property text vector and the industry information vector; In response to the semantic similarity exceeding a preset similarity threshold, the enterprise technology points of the target enterprise are determined.

7. A corporate credit processing device, characterized in that, include: The data acquisition module is used to acquire publicly available financial data, publicly available intellectual property data, and industry technology information of the target company's target industry. The enterprise technology point determination module is used to determine at least one enterprise technology point of the target enterprise based on the publicly available intellectual property data and the industry technology information. The attention weight determination module is used to determine the attention weight of each of the enterprise's technical points relative to each technical hotspot in the industry technical information, based on the publicly available intellectual property data and each of the enterprise's technical points. The credit granting result determination module is used to determine the credit granting result of the target enterprise based on the publicly available financial data, the technical points of each enterprise, and the attention weight.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the enterprise credit processing method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the enterprise credit processing method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the enterprise credit processing method according to any one of claims 1-6.

Images