Enterprise data processing method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides an enterprise data processing method and device, electronic equipment and a storage medium, and belongs to the technical field of financial science and technology. The method comprises the following steps: acquiring multi-dimensional enterprise data of a plurality of candidate enterprises; preprocessing the multi-dimensional enterprise data to obtain target enterprise data; performing feature extraction on the target enterprise data to obtain multi-dimensional enterprise features; market change information and enterprise demand information are obtained, and the screening rule weight of the screening model is adjusted according to the market change information and the enterprise demand information to obtain a target screening model; performing matching scoring on each multi-dimensional enterprise feature through a target screening model to obtain an enterprise matching scoring result; and determining a target recommended enterprise from the plurality of candidate enterprises according to an enterprise matching scoring result. Based on this, according to the embodiment of the invention, the enterprise multi-dimensional data is integrated, analyzed and processed, and the customer obtaining strategy is dynamically adjusted according to the market change and the enterprise need, so that the customer obtaining efficiency and quality can be remarkably improved.

Description

Enterprise data processing methods and apparatus, electronic devices and storage media Technical Field

[0001] This application relates to the field of financial technology (Fintech), and more particularly to an enterprise data processing method and apparatus, electronic device and storage medium. Background Technology

[0002] Currently, while some technologies and products exist in the market for bulk customer acquisition and target customer screening, existing technologies typically rely on a single data source (such as publicly available business information or third-party databases), making it difficult to comprehensively cover a company's multi-dimensional information (such as funding needs and operational status). Furthermore, insufficient cross-platform data integration capabilities easily lead to data silos, affecting the accuracy of customer acquisition. Moreover, in the process of bulk customer acquisition, there is often a lack of real-time awareness and dynamic adjustment capabilities regarding market changes and company needs. For example, failure to promptly detect adjustments in market policies, changes in company funding needs, or changes in operational status results in outdated customer acquisition strategies, negatively impacting both efficiency and quality. Summary of the Invention

[0003] The main objective of this application is to propose an enterprise data processing method, apparatus, electronic device, and storage medium. By integrating and analyzing enterprise multi-dimensional data and dynamically adjusting customer acquisition strategies according to market changes and enterprise needs, the efficiency and quality of customer acquisition can be significantly improved, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0004] To achieve the above objectives, a first aspect of this application proposes an enterprise data processing method, the method comprising:

[0005] The process involves: acquiring multi-dimensional enterprise data from multiple candidate companies; preprocessing the multi-dimensional enterprise data to obtain target company data; extracting features from the target company data to obtain multi-dimensional enterprise features; acquiring market change information and enterprise demand information; adjusting the weights of the screening rules of the screening model based on the market change information and the enterprise demand information to obtain a target screening model; matching and scoring each of the multi-dimensional enterprise features using the target screening model to obtain enterprise matching score results; and determining the target recommended company from the multiple candidate companies based on the enterprise matching score results.

[0006] In some embodiments, the multi-dimensional enterprise data includes structured data and unstructured data. The process of obtaining multi-dimensional enterprise data from multiple candidate enterprises includes: obtaining the structured data of the multiple candidate enterprises through an application programming interface (API), wherein the structured data includes enterprise operating data, movable asset financing data, and business registration data; and obtaining the unstructured data of the multiple candidate enterprises through a web crawler, wherein the unstructured data includes news website data, social media data, and policy document data.

[0007] In some embodiments, the preprocessing of the multi-dimensional enterprise data to obtain target enterprise data includes: de-identifying the multi-dimensional enterprise data to obtain de-identified enterprise data; and cleaning and standardizing the de-identified enterprise data to obtain target enterprise data.

[0008] In some embodiments, the step of extracting features from the target enterprise data to obtain multi-dimensional enterprise features includes: extracting features from the structured data to obtain structured features; extracting features from the unstructured data to obtain unstructured features; determining the feature weights of the structured features and the unstructured features based on a weighted scoring method; and concatenating the structured features and the unstructured features according to the feature weights to obtain multi-dimensional enterprise features.

[0009] In some embodiments, adjusting the screening rule weights of the screening model based on the market change information and the enterprise demand information to obtain a target screening model includes: updating the first target model parameters of the screening model through reinforcement learning in response to the market change information meeting a first preset condition; updating the second target model parameters of the screening model through reinforcement learning in response to the enterprise demand information meeting a second preset condition; and adjusting the screening rule weights of the screening model based on the first target model parameters and the second target model parameters to obtain the target screening model.

[0010] In some embodiments, the method further includes: acquiring marketing feedback data; determining a marketing conversion rate based on the marketing feedback data; and updating the model parameters of the target selection model through incremental learning in response to the marketing conversion rate being lower than a preset threshold.

[0011] In some embodiments, the method further includes: constructing a corporate profile of the target recommended enterprise based on the multi-dimensional corporate characteristics; obtaining multiple candidate financial products and extracting features from the candidate financial products to obtain financial product features; matching the corporate profile with the financial product features to obtain a target financial product; and recommending the target financial product to the target recommended enterprise; or, extracting risk features from the corporate profile, inputting the risk features into a risk model to generate a risk score for the target recommended enterprise; or, extracting industry features and financial features from the corporate profile, clustering the industry features and the financial features to obtain clustering results, and adjusting the product recommendation strategy for marketing to the target recommended enterprise based on the clustering results.

[0012] To achieve the above objectives, a second aspect of this application provides an enterprise data processing apparatus, comprising: an acquisition module for acquiring multi-dimensional enterprise data of multiple candidate enterprises; a preprocessing module for preprocessing the multi-dimensional enterprise data to obtain target enterprise data; an extraction module for extracting features from the target enterprise data to obtain multi-dimensional enterprise features; an adjustment module for acquiring market change information and enterprise demand information, and adjusting the weights of the screening rules of a screening model according to the market change information and the enterprise demand information to obtain a target screening model; a matching module for matching and scoring each of the multi-dimensional enterprise features through the target screening model to obtain an enterprise matching score result; and a determination module for determining a target recommended enterprise from the multiple candidate enterprises based on the enterprise matching score result.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] This application proposes a method, apparatus, electronic device, and storage medium for processing enterprise data. The method involves acquiring multi-dimensional enterprise data from multiple candidate enterprises; preprocessing the multi-dimensional enterprise data to obtain target enterprise data; extracting features from the target enterprise data to obtain multi-dimensional enterprise features; acquiring market change information and enterprise demand information; adjusting the weights of the screening rules of the screening model based on these information to obtain a target screening model; matching and scoring each multi-dimensional enterprise feature using the target screening model to obtain enterprise matching scores; and determining the target recommended enterprise from multiple candidate enterprises based on the enterprise matching scores. By acquiring multi-dimensional enterprise data from multiple candidate enterprises, preprocessing the multi-dimensional enterprise data to obtain target enterprise data, and extracting features from the target enterprise data, this application not only extracts multi-dimensional enterprise features that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate enterprises, but also dynamically adjusts the screening model based on market change information and enterprise demand information to obtain a target screening model. This dynamically adjusts the customer acquisition strategy and improves the accuracy of the screening. Then, a target screening model is used to match and score the multi-dimensional characteristics of each candidate company. By scoring the matching degree of the candidate companies, high-quality, high-intent target recommendation companies are prioritized from multiple candidate companies based on the matching score results, so that marketing operations can be carried out on the target recommendation companies in the future. This approach can not only improve marketing conversion rate and reduce customer acquisition cost, but also significantly improve the efficiency and quality of customer acquisition. Based on this, the embodiments of this application, by integrating and analyzing multi-dimensional data of enterprises and dynamically adjusting customer acquisition strategies according to market changes and enterprise needs, can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness. Attached Figure Description

[0016] Figure 1 is a flowchart of the enterprise data processing method provided in the embodiment of this application; Figure 2 is a flowchart of step S101 in Figure 1; Figure 3 is a flowchart of step S102 in Figure 1; Figure 4 is a flowchart of step S103 in Figure 1; Figure 5 is a flowchart of step S104 in Figure 1; Figure 6 is a flowchart of steps S601 to S603; Figure 7 is a flowchart of steps S701 to S702; Figure 8 is a structural schematic diagram of the enterprise data processing device provided in the embodiment of this application; Figure 9 is a hardware structural schematic diagram of the electronic device provided in the embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] First, let's clarify some terms used in this application: Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. AI also refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0021] Movable asset financing data: Movable asset financing refers to the act of an enterprise or individual using their own movable assets (such as inventory, accounts receivable, equipment, machinery, raw materials, etc.) as collateral to obtain financial support from banks or other financial institutions. Movable asset financing data encompasses various information and data generated around this financing activity. This data mainly includes: collateral information (such as type, quantity, value, etc.), financing entity information (financing party and fund provider), financing transaction information (amount, term, interest rate, etc.), ownership registration information (encumbrances on the collateral), and regulatory information (real-time monitoring data of the collateral).

[0022] China Central Depository & Clearing Co., Ltd. (CCDC) is an official platform established by the Credit Reference Center of the People's Bank of China. CCDC is an important channel for obtaining data on movable asset financing.

[0023] Currently, while some technologies and products exist in the market for bulk customer acquisition and target customer screening, existing technologies typically rely on a single data source (such as publicly available business information or third-party databases), making it difficult to comprehensively cover a company's multi-dimensional information (such as funding needs and operational status). Furthermore, insufficient cross-platform data integration capabilities easily lead to data silos, affecting the accuracy of customer acquisition. Moreover, in the process of bulk customer acquisition, there is often a lack of real-time awareness and dynamic adjustment capabilities regarding market changes and company needs. For example, failure to promptly detect adjustments in market policies, changes in company funding needs, or changes in operational status results in outdated customer acquisition strategies, negatively impacting both efficiency and quality.

[0024] Based on this, embodiments of this application provide a method, apparatus, electronic device, and storage medium for processing enterprise data. The method involves acquiring multi-dimensional enterprise data from multiple candidate enterprises; preprocessing the multi-dimensional enterprise data to obtain target enterprise data; extracting features from the target enterprise data to obtain multi-dimensional enterprise features; acquiring market change information and enterprise demand information; adjusting the weights of the screening rules of the screening model based on the market change information and enterprise demand information to obtain a target screening model; matching and scoring each multi-dimensional enterprise feature using the target screening model to obtain an enterprise matching score result; and determining the target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result. By acquiring multi-dimensional enterprise data from multiple candidate enterprises, preprocessing the multi-dimensional enterprise data to obtain target enterprise data, and extracting features from the target enterprise data to obtain multi-dimensional enterprise features, this embodiment of the application not only extracts multi-dimensional enterprise features that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate enterprises, but also dynamically adjusts the screening model based on market change information and enterprise demand information to obtain a target screening model, thereby dynamically adjusting customer acquisition strategies and improving the accuracy of screening. Then, a target screening model is used to match and score the multi-dimensional characteristics of each candidate company. By scoring the matching degree of the candidate companies, high-quality, high-intent target recommendation companies are prioritized from multiple candidate companies based on the matching score results, so that marketing operations can be carried out on the target recommendation companies in the future. This approach can not only improve marketing conversion rate and reduce customer acquisition cost, but also significantly improve the efficiency and quality of customer acquisition. Based on this, the embodiments of this application, by integrating and analyzing multi-dimensional data of enterprises and dynamically adjusting customer acquisition strategies according to market changes and enterprise needs, can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0025] The enterprise data processing method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the enterprise data processing method in this application is described.

[0026] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0027] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0028] The enterprise data processing method provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the enterprise data processing method, but is not limited to the above forms.

[0029] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0030] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0031] Figure 1 is an optional flowchart of an enterprise data processing method provided in an embodiment of this application. The method in Figure 1 may include, but is not limited to, steps S101 to S106.

[0032] Step S101: Obtain multi-dimensional enterprise data from multiple candidate enterprises; Step S102: Preprocess the multi-dimensional enterprise data to obtain target enterprise data; Step S103: Extract features from the target enterprise data to obtain multi-dimensional enterprise features; Step S104: Obtain market change information and enterprise demand information, and adjust the weights of the screening rules of the screening model according to the market change information and enterprise demand information to obtain the target screening model; Step S105: Match and score each multi-dimensional enterprise feature using the target screening model to obtain the enterprise matching score result; Step S106: Determine the target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result.

[0033] In step S101 of some embodiments, multi-dimensional enterprise data of multiple candidate companies is acquired. Classified by data structure type, multi-dimensional enterprise data includes structured and unstructured data. Structured data includes, but is not limited to, enterprise operating data, movable asset financing data, and business registration data related to the candidate companies. Enterprise operating data includes financial data, such as net profit, debt-to-asset ratio, and cash flow, which can be obtained from the enterprise annual report database. Movable asset financing data includes financing amount, financing announcement time, and historical financing success rate, which can be obtained from the China Securities Depository and Clearing Corporation Limited (CSDC). Business registration data includes registered capital, years of establishment, and shareholder structure, which can be obtained from the business registration database. Unstructured data includes, but is not limited to, news website data, social media data, and policy document data related to the candidate companies. News website data includes enterprise announcements and industry dynamics data; social media data includes enterprise-related topic discussion data; and policy document data includes industry support policy data. By integrating multi-source data, multi-dimensional enterprise data of candidate companies is obtained to avoid the impact of a single data source on the accuracy of customer acquisition.

[0034] In step S102 of some embodiments, the multi-dimensional enterprise data is preprocessed to obtain the target enterprise data. The acquired multi-dimensional enterprise data may contain sensitive data involving personal privacy and trade secrets, and may also contain missing fields, misaligned fields, or non-standardized data. Therefore, it is necessary to preprocess the multi-dimensional enterprise data. Preprocessing includes, but is not limited to, data anonymization, data cleaning, and data standardization to facilitate subsequent feature extraction.

[0035] In step S103 of some embodiments, feature extraction is performed on the target enterprise data to obtain multi-dimensional enterprise features. Feature extraction is performed on structured data to obtain structured features; feature extraction is performed on unstructured data to obtain unstructured features; the feature weights of the structured and unstructured features are determined based on a weighted scoring method; the structured and unstructured features are concatenated according to the feature weights to obtain multi-dimensional enterprise features, so as to more comprehensively reflect the enterprise profile and improve the accuracy of enterprise customer screening.

[0036] In step S104 of some embodiments, market change information and enterprise demand information are obtained, and the weights of the screening rules of the screening model are adjusted according to the market change information and enterprise demand information to obtain the target screening model. This application embodiment introduces a screening model with a dynamic adjustment mechanism. The screening model includes, but is not limited to, the DeepSeek rule model, which can update and adjust the weights of the screening rules in real time according to market change information and enterprise demand information. In other words, it can dynamically adjust the customer acquisition strategy based on real-time market changes and enterprise demands, enabling the adjusted target screening model to quickly respond to the latest market changes and customer needs, thus helping to improve customer acquisition efficiency and quality.

[0037] In step S105 of some embodiments, a target screening model is used to match and score various multi-dimensional enterprise characteristics to obtain enterprise matching score results. The target screening model, a dynamically optimized model based on multi-dimensional enterprise data quality assessment, quantifies multi-dimensional enterprise characteristics (such as operational stability, credit history, and matching degree of funding needs). Lower scores indicate higher quality. In the enterprise matching score results, low-risk, high-potential enterprise clients are classified as low-scoring clients, representing a high degree of matching with target client characteristics (e.g., clear funding needs and stable operations); conversely, high-risk, low-potential enterprise clients are classified as high-scoring clients, representing a low degree of matching with target client characteristics (e.g., no funding needs and poor credit). By calculating the matching degree of the multi-dimensional enterprise characteristics of each candidate enterprise, the enterprise matching score results for each candidate enterprise are obtained. Therefore, using a target screening model to match and score various multi-dimensional enterprise characteristics can improve the comprehensiveness of the scoring, thereby effectively screening out high-quality enterprises.

[0038] In step S106 of some embodiments, a target recommended enterprise is determined from multiple candidate enterprises based on the enterprise matching score results. For example, based on the enterprise matching score results, high-match-degree enterprise customers and low-match-degree enterprise customers can be determined according to the matching degree stratification, and high-match-degree enterprise customers can be selected and output as target recommended enterprises. Marketing operations can then be carried out on these target recommended enterprises.

[0039] Steps S101 to S106 of this embodiment involve: acquiring multi-dimensional enterprise data from multiple candidate companies; preprocessing the multi-dimensional enterprise data to obtain target company data; extracting features from the target company data to obtain multi-dimensional enterprise features; acquiring market change information and enterprise demand information; adjusting the weights of the screening rules of the screening model based on the market change information and enterprise demand information to obtain a target screening model; matching and scoring each multi-dimensional enterprise feature using the target screening model to obtain an enterprise matching score result; and determining the target recommended company from multiple candidate companies based on the enterprise matching score result. This embodiment not only extracts multi-dimensional enterprise features that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate companies, but also dynamically adjusts the screening model based on market change information and enterprise demand information to obtain a target screening model, thereby dynamically adjusting the customer acquisition strategy and improving the accuracy of screening. Then, a target screening model is used to match and score the multi-dimensional characteristics of each candidate company. By scoring the matching degree of the candidate companies, high-quality, high-intent target recommendation companies are prioritized from multiple candidate companies based on the matching score results, so that marketing operations can be carried out on the target recommendation companies in the future. This approach can not only improve marketing conversion rate and reduce customer acquisition cost, but also significantly improve the efficiency and quality of customer acquisition. Based on this, the embodiments of this application, by integrating and analyzing multi-dimensional data of enterprises and dynamically adjusting customer acquisition strategies according to market changes and enterprise needs, can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0040] Please refer to Figure 2. In some embodiments, step S101 may include, but is not limited to, steps S201 to S202: Step S201: Obtain structured data of multiple candidate companies through an application programming interface, wherein the structured data includes enterprise operation data, movable property financing data, and business registration data; Step S202: Obtain unstructured data of multiple candidate companies through a web crawler, wherein the unstructured data includes news website data, social media data, and policy document data.

[0041] Understandably, multi-dimensional enterprise data includes both structured and unstructured data. Structured data from multiple candidate companies can be obtained via Application Programming Interface (API) databases, such as enterprise annual report databases, China Securities Depository and Clearing Corporation Limited (CSDC) databases, and industrial and commercial registration databases. This structured data includes enterprise operating data, movable asset financing data, and business registration data. Unstructured data from multiple candidate companies can be crawled via web crawlers, including data from news websites, social media, and policy documents. Based on this, APIs and web crawlers can automate the batch acquisition and processing of enterprise information, significantly improving customer acquisition efficiency and overcoming the inefficiency and errors of traditional manual processing methods.

[0042] Please refer to Figure 3. In some embodiments, step S102 may include, but is not limited to, steps S301 to S302: Step S301, de-identify the multi-dimensional enterprise data to obtain de-identified enterprise data; Step S302, clean and standardize the de-identified enterprise data to obtain target enterprise data.

[0043] Understandably, the multi-dimensional enterprise data acquired may contain sensitive information such as personal privacy and trade secrets. Therefore, data anonymization techniques can be used to anonymize this data, obtaining anonymized enterprise data. For example, fields involving personal privacy (such as names and ID numbers) and trade secrets (such as financial data and customer transaction records) can be anonymized. Similarly, for non-sensitive but protected data (such as company addresses and contact information), partial anonymization can be performed based on business needs. Specific anonymization methods include, but are not limited to, hashing, range fuzzing, and keyword replacement. This approach avoids the risk of privacy leaks, ensuring compliance and security of data usage while improving customer acquisition efficiency and quality.

[0044] Understandably, the acquired multi-dimensional enterprise data may contain missing fields, misaligned fields, or lack of standardization. Therefore, it is necessary to clean and standardize the anonymized enterprise data to obtain the target enterprise data. For example, for field alignment, the unified social credit code or enterprise name can be used as the primary key to align the three types of data: enterprise operating data, movable asset financing data, and business registration data. For handling missing fields, interpolation algorithms (such as KNN interpolation) or rule-based filling can be used to fill in missing fields, such as marking missing debt-to-asset ratios as low credit risk.

[0045] Please refer to Figure 4. In some embodiments, step S103 may include, but is not limited to, steps S401 to S402: Step S401, extracting features from structured data to obtain structured features; Step S402, extracting features from unstructured data to obtain unstructured features; Step S403, determining the feature weights of structured and unstructured features based on a weighted scoring method; Step S404, concatenating structured and unstructured features according to their feature weights to obtain multi-dimensional enterprise features.

[0046] Understandably, feature extraction from structured data, including enterprise operating data, movable asset financing data, and business registration data, can yield structured features, such as financial features, industry features, and risk features.

[0047] Understandably, feature extraction from unstructured data, including news website data, social media data, and policy document data, can yield unstructured features, such as public opinion features, industry-related features, and behavioral features.

[0048] Understandably, different feature weights can be assigned to different structured and unstructured features based on a weighted scoring method, such as a feature weight of 0.2 for financial features and 0.3 for public opinion features. By concatenating structured and unstructured features according to these weights, multi-dimensional enterprise features can be obtained, providing a more comprehensive picture of the enterprise and improving the accuracy of customer screening.

[0049] Please refer to Figure 5. In some embodiments, step S104 may include, but is not limited to, steps S501 to S503: Step S501, in response to market change information satisfying a first preset condition, the first target model parameters of the screening model are updated through reinforcement learning; Step S502, in response to enterprise demand information satisfying a second preset condition, the second target model parameters of the screening model are updated through reinforcement learning; Step S503, the screening rule weights of the screening model are adjusted according to the first target model parameters and the second target model parameters to obtain the target screening model.

[0050] Understandably, when changes in market information are detected, for example, when the favorableness of industry policies declines, such as the tightening of "green finance" policies, with the favorableness dropping from 70% to 50%, meeting the first preset condition (e.g., a decrease in favorableness exceeding 10%), the first target model parameter "green finance recommendation" of the screening model is updated through reinforcement learning, triggering an adjustment in the weight of the screening rule "green finance recommendation" of the screening model (e.g., from 0.7 to 0.4). Similarly, when economic indicators, such as GDP growth falling below 5% and failing to meet expectations, meet the first preset condition (e.g., a threshold of 5%), the first target model parameter "cash flow stability" of the screening model is updated through reinforcement learning, triggering an adjustment in the weight of the screening rule "cash flow stability" of the screening model (from 0.3 to 0.6).

[0051] Understandably, when changes in enterprise demand information are detected, for example, when the customer conversion rate of a certain type of financial product (such as "green credit") is 8%, meeting the second preset condition (such as being below the threshold of 10%), the weight of the screening rule "industry policy benefit" in the screening model is adjusted, such as reducing the weight of "industry policy benefit", or the screening rule of the screening model is replaced, such as deleting the old rule and adding a new rule "industry policy benefit ≤ 30% → recommend 'traditional credit products'".

[0052] Please refer to Figure 6. In some embodiments, the enterprise data processing method of this application may include, but is not limited to, steps S601 to S603: Step S601, obtaining marketing feedback data; Step S602, determining the marketing conversion rate based on the marketing feedback data; Step S603, in response to the marketing conversion rate being lower than a preset threshold, updating the model parameters of the target screening model through incremental learning.

[0053] It is understood that the enterprise data processing method in this application embodiment, in addition to adjusting the weights of the screening rules of the screening model based on market change information and enterprise demand information to obtain the target screening model, can further optimize the model parameters of the target screening model based on marketing results. Specifically, by acquiring marketing feedback data and analyzing it to determine the marketing conversion rate, when the marketing conversion rate is detected to be lower than a preset threshold, the model parameters of the target screening model can be updated through incremental learning to achieve retraining of the target screening model. For example, online learning algorithms can be used to update the model parameters of the target screening model, such as fine-tuning the target screening model after adding public opinion data. Based on this, this application embodiment adopts a real-time feedback mechanism, which can dynamically optimize the model parameters of the screening model based on marketing feedback data (such as conversion rate, customer feedback, etc.), thereby improving customer acquisition efficiency.

[0054] Please refer to Figure 7. In some embodiments, the enterprise data processing method of this application may include, but is not limited to, steps S701 to S702: Step S701, constructing an enterprise profile of the target recommended enterprise based on multi-dimensional enterprise characteristics; Step S702, obtaining multiple candidate financial products, extracting features from the candidate financial products to obtain financial product features, matching the enterprise profile with the financial product features to obtain the target financial product, and recommending the target financial product to the target recommended enterprise; or, extracting risk features from the enterprise profile, inputting the risk features into a risk model to generate a risk score for the target recommended enterprise; or, extracting industry features and financial features from the enterprise profile, clustering the industry features and financial features to obtain clustering results, and adjusting the product recommendation strategy for marketing to the target recommended enterprise based on the clustering results.

[0055] It is understood that the enterprise data processing method in this application embodiment can efficiently filter out high-quality, highly willing target recommended enterprises. After filtering out the target recommended enterprises, marketing operations can be carried out on the target recommended enterprises, including but not limited to promoting financial products, B2B sales (such as supply chain services, enterprise software subscriptions), and marketing (such as customized advertising). For example, an enterprise profile of the target recommended enterprise can be constructed based on multi-dimensional enterprise characteristics. The enterprise profile can be matched with the characteristics of candidate financial products (such as loan amount, interest rate, and term) in multiple dimensions to obtain the target financial product, and the target financial product can be recommended to the target recommended enterprise.

[0056] Understandably, product recommendation strategies for marketing to target companies can also be adjusted. For example, industry and financial characteristics can be extracted from the company profiles of target companies, and the K-means algorithm can be used to cluster these characteristics. The clustering results include different customer groups such as high-growth technology companies and traditional manufacturing companies. Based on the clustering results, the product recommendation strategy for marketing to high-growth technology companies can be automatically adjusted and optimized. For example, a "fast-track approval channel" can be added for high-growth technology companies.

[0057] Understandably, risk assessments can also be conducted on target recommended companies. Risk characteristics are extracted from the company profile, input into a risk model, and a risk score is generated for the target recommended company. For example, risk characteristics such as "number of legal disputes," "negative keywords in public opinion," and "cash flow stability" from the company profile can be input into the risk model to generate a risk score. When the risk score exceeds a risk threshold, a risk warning is triggered and the credit limit is automatically reduced.

[0058] The enterprise data processing method of this application is further illustrated below with reference to specific embodiments.

[0059] Example 1: Enterprise comprehensive scoring model based on multi-source data. Scenario: A bank needs to screen potential loan customers.

[0060] The data from multiple sources is integrated as follows: Business registration data: registered capital (≥10 million RMB, low score), years of establishment (≥5 years, low score), business scope (matching the loan product, low score).

[0061] Data from China Securities Depository and Clearing Corporation (CSDC): Does the company have recent financing needs? (Lower score if yes) and historical financing success rate (lower score if high success rate).

[0062] DeepSeek rule model: Analyze corporate public opinion through natural language processing (such as keywords like "expansion plan" and "tight funds," with negative keywords scoring high).

[0063] The scoring logic is as follows: Weighted formula: Overall score = 0.3 Credit score +0.2 Business performance score +0.2 Funding needs matching score +0.1 Public opinion score +0.2 Scores in other dimensions.

[0064] The following is a scoring example: Company A: Registered capital of 15 million (low score), established for 8 years (low score), business scope matches (low score), has recent financing needs (low score), no negative public opinion (low score) → Overall score: 20 points (high-quality customer).

[0065] Company B: Registered capital of 2 million (high score), established for 1 year (high score), mismatched business scope (high score), no financing needs (high score), public opinion includes "tight funds" (high score) → Overall score: 80 points (low quality customer).

[0066] Example 2: Score change scenario under dynamic adjustment mechanism: A certain industry experiences a surge in funding demand due to favorable policies (such as new energy subsidies).

[0067] The dynamic adjustment logic is as follows: Triggering condition: The system detects the keyword "new energy subsidy" in policy documents, and the industry's funding demand data increases by 30% year-on-year.

[0068] The DeepSeek rule model has been adjusted as follows: the weight of "matching degree of capital needs" for enterprises in this industry has been reduced (from 0.2 to 0.3); the dimension of "relevance of policy benefits" has been added (e.g., whether the enterprise is involved in new energy business, the higher the relevance, the lower the score).

[0069] The scoring example changes are as follows: Company C (new energy company): Overall score before adjustment: 60 points (medium quality); Overall score after adjustment: 40 points (high-quality customer, due to improved matching of policy benefits and funding needs).

[0070] Company D (Traditional Manufacturing): Overall score before adjustment: 50 points (high quality); Overall score after adjustment: 65 points (due to irrelevant policy benefits and decreased matching degree of capital needs).

[0071] This application not only acquires multi-dimensional enterprise data from multiple candidate companies, preprocesses the multi-dimensional enterprise data to obtain target company data, extracts features from the target company data to obtain multi-dimensional enterprise features, acquires market change information and enterprise demand information, adjusts the weights of the screening rules of the screening model based on the market change information and enterprise demand information to obtain a target screening model, matches and scores each multi-dimensional enterprise feature using the target screening model to obtain enterprise matching scores, and determines the target recommended company from multiple candidate companies based on the enterprise matching scores. This embodiment not only extracts multi-dimensional enterprise features that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate companies, but also dynamically adjusts the screening model based on market change information and enterprise demand information to obtain a target screening model, thereby dynamically adjusting customer acquisition strategies and improving the accuracy of screening. Then, a target selection model is used to match and score the multi-dimensional characteristics of each candidate company. By scoring the candidate companies based on their matching degree, high-quality, high-intent target companies are prioritized from multiple candidate companies for subsequent marketing operations. This approach not only improves marketing conversion rates and reduces customer acquisition costs, but also significantly improves the efficiency and quality of customer acquisition.

[0072] Based on this, the embodiments of this application integrate and analyze enterprise data from multiple dimensions, and dynamically adjust customer acquisition strategies according to market changes and enterprise needs, which can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0073] Referring to Figure 8, this application embodiment also provides an enterprise data processing apparatus that can implement the above-described enterprise data processing method. The apparatus includes: an acquisition module 810 for acquiring multi-dimensional enterprise data of multiple candidate enterprises; a preprocessing module 820 for preprocessing the multi-dimensional enterprise data to obtain target enterprise data; an extraction module 830 for extracting features from the target enterprise data to obtain multi-dimensional enterprise features; an adjustment module 840 for acquiring market change information and enterprise demand information, and adjusting the weights of the screening rules of the screening model according to the market change information and enterprise demand information to obtain a target screening model; a matching module 850 for matching and scoring each multi-dimensional enterprise feature through the target screening model to obtain an enterprise matching score result; and a determination module 860 for determining a target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result.

[0074] In some embodiments of this application, the acquisition module 810 acquires multi-dimensional enterprise data of multiple candidate enterprises; the preprocessing module 820 preprocesses the multi-dimensional enterprise data to obtain target enterprise data; the extraction module 830 extracts features from the target enterprise data to obtain multi-dimensional enterprise features; the adjustment module 840 acquires market change information and enterprise demand information, and adjusts the weights of the screening rules of the screening model according to the market change information and enterprise demand information to obtain the target screening model; the matching module 850 performs matching scoring on each multi-dimensional enterprise feature through the target screening model to obtain the enterprise matching score result; and the determination module 860 determines the target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result.

[0075] In some embodiments of this application, multi-dimensional enterprise data from multiple candidate companies is obtained. Classified by data structure type, multi-dimensional enterprise data includes structured and unstructured data. Structured data includes, but is not limited to, enterprise operating data, movable asset financing data, and business registration data related to the candidate companies. Enterprise operating data includes financial data, such as net profit, debt-to-asset ratio, and cash flow, which can be obtained from enterprise annual report databases. Movable asset financing data includes financing amount, financing announcement time, and historical financing success rate, which can be obtained from the China Securities Depository and Clearing Corporation Limited (CSDC). Business registration data includes registered capital, years of establishment, and shareholder structure, which can be obtained from business registration databases. Unstructured data includes, but is not limited to, news website data, social media data, and policy document data related to the candidate companies. News website data includes enterprise announcements and industry dynamics data; social media data includes enterprise-related topic discussions; and policy document data includes industry support policy data. By integrating multi-source data, multi-dimensional enterprise data of candidate companies is obtained to avoid the impact of a single data source on the accuracy of customer acquisition.

[0076] In some embodiments of this application, multi-dimensional enterprise data is preprocessed to obtain target enterprise data. The acquired multi-dimensional enterprise data may contain sensitive data involving personal privacy and trade secrets, and may also contain missing fields, misaligned fields, or non-standardized data. Therefore, preprocessing of the multi-dimensional enterprise data is necessary. Preprocessing includes, but is not limited to, data anonymization, data cleaning, and data standardization to facilitate subsequent feature extraction.

[0077] In some embodiments of this application, feature extraction is performed on target enterprise data to obtain multi-dimensional enterprise features. Feature extraction is performed on structured data to obtain structured features; feature extraction is performed on unstructured data to obtain unstructured features; the feature weights of the structured and unstructured features are determined based on a weighted scoring method; the structured and unstructured features are concatenated according to the feature weights to obtain multi-dimensional enterprise features, so as to more comprehensively reflect the enterprise profile and improve the accuracy of enterprise customer screening.

[0078] In some embodiments of this application, market change information and enterprise demand information are acquired, and the weights of the screening rules of the screening model are adjusted according to the market change information and enterprise demand information to obtain the target screening model. Embodiments of this application introduce a screening model with a dynamic adjustment mechanism. The screening model includes, but is not limited to, the DeepSeek rule model, which can update and adjust the weights of the screening rules in real time according to market change information and enterprise demand information. In other words, it can dynamically adjust the customer acquisition strategy based on real-time market changes and enterprise needs, enabling the adjusted target screening model to quickly respond to the latest market changes and customer needs, thus helping to improve customer acquisition efficiency and quality.

[0079] In some embodiments of this application, a target screening model is used to match and score various multi-dimensional enterprise characteristics, resulting in enterprise matching scores. The target screening model, a dynamically optimized model based on multi-dimensional enterprise data quality assessment, quantifies multi-dimensional enterprise characteristics (such as operational stability, credit history, and matching degree of funding needs). Lower scores indicate higher quality. In the enterprise matching scores, low-risk, high-potential enterprise clients are classified as low-scoring clients, representing a high degree of matching with the target client characteristics (e.g., clear funding needs and stable operations); conversely, high-risk, low-potential enterprise clients are classified as high-scoring clients, representing a low degree of matching with the target client characteristics (e.g., no funding needs and poor credit). By calculating the matching degree of the multi-dimensional enterprise characteristics of each candidate enterprise, the enterprise matching scores for each candidate enterprise are obtained. Therefore, using a target screening model to match and score various multi-dimensional enterprise characteristics can improve the comprehensiveness of the scoring, thereby effectively screening out high-quality enterprises.

[0080] In some embodiments of this application, target recommended enterprises are determined from multiple candidate enterprises based on enterprise matching score results. For example, based on the enterprise matching score results, high-match-degree enterprise customers and low-match-degree enterprise customers can be identified in a stratified manner, and high-match-degree enterprise customers can be selected as target recommended enterprises. Subsequently, marketing operations can be carried out on these target recommended enterprises.

[0081] Based on this, the enterprise data processing apparatus of this application embodiment includes an acquisition module 810 acquiring multi-dimensional enterprise data of multiple candidate enterprises; a preprocessing module 820 preprocessing the multi-dimensional enterprise data to obtain target enterprise data; an extraction module 830 extracting features from the target enterprise data to obtain multi-dimensional enterprise features; an adjustment module 840 acquiring market change information and enterprise demand information, and adjusting the weights of the screening rules of the screening model according to the market change information and enterprise demand information to obtain a target screening model; a matching module 850 matching and scoring each multi-dimensional enterprise feature through the target screening model to obtain an enterprise matching score result; and a determination module 860 determining the target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result. This application acquires multi-dimensional enterprise data from multiple candidate companies; preprocesses the multi-dimensional enterprise data to obtain target company data; extracts features from the target company data to obtain multi-dimensional enterprise features; acquires market change information and enterprise demand information, and adjusts the weights of the screening rules of the screening model based on the market change information and enterprise demand information to obtain a target screening model; matches and scores each multi-dimensional enterprise feature using the target screening model to obtain enterprise matching scores; and determines the target recommended company from multiple candidate companies based on the enterprise matching scores. By acquiring multi-dimensional enterprise data from multiple candidate companies, preprocessing the multi-dimensional enterprise data to obtain target company data, and extracting features from the target company data to obtain multi-dimensional enterprise features, this application not only extracts multi-dimensional enterprise features that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate companies, but also dynamically adjusts the screening model based on market change information and enterprise demand information to obtain a target screening model, thereby dynamically adjusting customer acquisition strategies and improving the accuracy of screening. Then, a target screening model is used to match and score the multi-dimensional characteristics of each candidate company. By scoring the matching degree of the candidate companies, high-quality, high-intent target recommendation companies are prioritized from multiple candidate companies based on the matching score results, so that marketing operations can be carried out on the target recommendation companies in the future. This approach can not only improve marketing conversion rate and reduce customer acquisition cost, but also significantly improve the efficiency and quality of customer acquisition. Based on this, the embodiments of this application, by integrating and analyzing multi-dimensional data of enterprises and dynamically adjusting customer acquisition strategies according to market changes and enterprise needs, can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0082] The specific implementation of the enterprise data processing device is basically the same as the specific implementation of the enterprise data processing method described above, and will not be repeated here.

[0083] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned enterprise data processing method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0084] Please refer to Figure 9, which illustrates the hardware structure of an electronic device according to another embodiment. The electronic device includes a processor 901, which can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute related programs to implement the technical solutions provided in the embodiments of this application.

[0085] The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and called by the processor 901 to execute the enterprise data processing method of the embodiments of this application. Specifically, this method involves: acquiring multi-dimensional enterprise data from multiple candidate enterprises; preprocessing the multi-dimensional enterprise data to obtain target enterprise data; extracting features from the target enterprise data to obtain multi-dimensional enterprise features; acquiring market change information and enterprise demand information; adjusting the weights of the screening rules of the screening model based on the market change information and enterprise demand information to obtain a target screening model; matching and scoring each multi-dimensional enterprise feature using the target screening model to obtain an enterprise matching score result; and determining the target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result. The process involves acquiring multi-dimensional enterprise data from multiple candidate enterprises, preprocessing the multi-dimensional enterprise data to obtain target enterprise data, and extracting features from the target enterprise data to obtain multi-dimensional enterprise features. This application not only extracts multi-dimensional enterprise characteristics that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate enterprises, but also dynamically adjusts the screening model based on market changes and enterprise needs to obtain a target screening model. This dynamically adjusts the customer acquisition strategy and improves the accuracy of screening. The target screening model then matches and scores the multi-dimensional enterprise characteristics of each candidate enterprise. By scoring the matching degree of candidate enterprises, high-quality, high-intent target recommendation enterprises are prioritized from multiple candidate enterprises based on the matching score results, so that subsequent marketing operations can be carried out on these target recommendation enterprises. This approach not only improves marketing conversion rates and reduces customer acquisition costs, but also significantly improves the efficiency and quality of customer acquisition. Based on this, this application, through the integrated analysis and processing of multi-dimensional enterprise data and the dynamic adjustment of customer acquisition strategies according to market changes and enterprise needs, can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0086] The input / output interface 903 is used to implement information input and output.

[0087] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0088] The bus transmits information between various components of the device, such as processor 901, memory 902, input / output interface 903, and communication interface 904.

[0089] The processor 901, memory 902, input / output interface 903, and communication interface 904 communicate with each other within the device via a bus.

[0090] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described enterprise data processing method.

[0091] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0092] The enterprise data processing method, apparatus, electronic device, and storage medium provided in this application not only acquire multi-dimensional enterprise data from multiple candidate enterprises; preprocess the multi-dimensional enterprise data to obtain target enterprise data; extract features from the target enterprise data to obtain multi-dimensional enterprise features; acquire market change information and enterprise demand information; adjust the weights of the screening rules of the screening model based on the market change information and enterprise demand information to obtain a target screening model; match and score each multi-dimensional enterprise feature using the target screening model to obtain an enterprise matching score result; and determine the target recommended enterprise from multiple candidate enterprises based on the enterprise matching score result. By acquiring multi-dimensional enterprise data from multiple candidate enterprises, preprocessing the multi-dimensional enterprise data to obtain target enterprise data, and extracting features from the target enterprise data to obtain multi-dimensional enterprise features, this application not only extracts multi-dimensional enterprise features that more comprehensively reflect the enterprise profile by collecting multi-dimensional enterprise data from candidate enterprises, but also dynamically adjusts the screening model based on market change information and enterprise demand information to obtain a target screening model, thereby dynamically adjusting customer acquisition strategies and improving the accuracy of screening. Then, a target screening model is used to match and score the multi-dimensional characteristics of each candidate company. By scoring the matching degree of the candidate companies, high-quality, high-intent target recommendation companies are prioritized from multiple candidate companies based on the matching score results, so that marketing operations can be carried out on the target recommendation companies in the future. This approach can not only improve marketing conversion rate and reduce customer acquisition cost, but also significantly improve the efficiency and quality of customer acquisition. Based on this, the embodiments of this application, by integrating and analyzing multi-dimensional data of enterprises and dynamically adjusting customer acquisition strategies according to market changes and enterprise needs, can significantly improve the efficiency and quality of customer acquisition, thereby quickly responding to customer needs and enhancing the enterprise's market competitiveness.

[0093] Those skilled in the art will understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable programs, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable programs, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0094] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0095] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0096] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0097] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0098] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification 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.

[0099] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0101] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0104] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for processing enterprise data, characterized in that, The method includes: acquiring multi-dimensional enterprise data from multiple candidate enterprises; preprocessing the multi-dimensional enterprise data to obtain target enterprise data; extracting features from the target enterprise data to obtain multi-dimensional enterprise features; acquiring market change information and enterprise demand information, adjusting the weights of the screening rules of the screening model according to the market change information and the enterprise demand information to obtain a target screening model; matching and scoring each of the multi-dimensional enterprise features using the target screening model to obtain an enterprise matching score result; and determining a target recommended enterprise from the multiple candidate enterprises based on the enterprise matching score result.

2. The method according to claim 1, characterized in that, The multi-dimensional enterprise data includes structured and unstructured data. Obtaining multi-dimensional enterprise data from multiple candidate enterprises includes: obtaining the structured data of multiple candidate enterprises through an application programming interface, wherein the structured data includes enterprise operating data, movable property financing data, and business registration data; and obtaining the unstructured data of multiple candidate enterprises through web crawlers, wherein the unstructured data includes news website data, social media data, and policy document data.

3. The method according to claim 1, characterized in that, The step of preprocessing the multi-dimensional enterprise data to obtain target enterprise data includes: de-identifying the multi-dimensional enterprise data to obtain de-identified enterprise data; and cleaning and standardizing the de-identified enterprise data to obtain target enterprise data.

4. The method according to claim 2, characterized in that, The step of extracting features from the target enterprise data to obtain multi-dimensional enterprise features includes: extracting features from the structured data to obtain structured features; extracting features from the unstructured data to obtain unstructured features; determining the feature weights of the structured features and the unstructured features based on a weighted scoring method; and concatenating the structured features and the unstructured features according to the feature weights to obtain multi-dimensional enterprise features.

5. The method according to claim 1, characterized in that, The step of adjusting the screening rule weights of the screening model based on the market change information and the enterprise demand information to obtain the target screening model includes: updating the first target model parameters of the screening model through reinforcement learning in response to the market change information meeting a first preset condition; updating the second target model parameters of the screening model through reinforcement learning in response to the enterprise demand information meeting a second preset condition; and adjusting the screening rule weights of the screening model based on the first target model parameters and the second target model parameters to obtain the target screening model.

6. The method according to claim 1, characterized in that, The method further includes: acquiring marketing feedback data; determining a marketing conversion rate based on the marketing feedback data; and updating the model parameters of the target selection model through incremental learning in response to the marketing conversion rate being lower than a preset threshold.

7. The method according to claim 1, characterized in that, The method further includes: constructing a corporate profile of the target recommended enterprise based on the multi-dimensional corporate characteristics; obtaining multiple candidate financial products and extracting features from the candidate financial products to obtain financial product features; matching the corporate profile with the financial product features to obtain a target financial product and recommending the target financial product to the target recommended enterprise; or, extracting risk features from the corporate profile, inputting the risk features into a risk model to generate a risk score for the target recommended enterprise; or, extracting industry features and financial features from the corporate profile, clustering the industry features and financial features to obtain clustering results, and adjusting the product recommendation strategy for marketing to the target recommended enterprise based on the clustering results.

8. An enterprise data processing device, characterized in that, The device includes: an acquisition module for acquiring multi-dimensional enterprise data of multiple candidate enterprises; a preprocessing module for preprocessing the multi-dimensional enterprise data to obtain target enterprise data; an extraction module for extracting features from the target enterprise data to obtain multi-dimensional enterprise features; an adjustment module for acquiring market change information and enterprise demand information, and adjusting the weights of the screening rules of the screening model according to the market change information and the enterprise demand information to obtain a target screening model; a matching module for matching and scoring each of the multi-dimensional enterprise features through the target screening model to obtain an enterprise matching score result; and a determination module for determining a target recommended enterprise from the multiple candidate enterprises based on the enterprise matching score result.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the enterprise data processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the enterprise data processing method according to any one of claims 1 to 7.

Images