Intelligent matching-based creditor right checking processing method for small, medium and micro-sized enterprises

Abstract

The invention provides an intelligent matching-based small, medium and micro-sized enterprise creditor's right checking processing method, which comprises the following steps of: firstly, building a creditor's right circulation intelligent platform, submitting applications and related materials by small, medium and micro-sized enterprises, verifying the authenticity of the creditor's right after the platform verifies the integrity, and completing three-party signature after the creditor's right transferee confirms; after the agreement takes effect, the creditor's right information is stored in a platform for evidence storage to form a tamper-resistant file, assets are screened in multiple dimensions based on the creditor's right information of the tamper-resistant file, a matching list is generated and pushed to a creditor's right transferring person, and a creditor's right management mechanism generates an electronic certificate corresponding to the file based on the creditor's right information in the tamper-resistant file and sends the electronic certificate to the creditor's right transferring person. After a creditor's right transferring person selects assets from the matching list, the platform carries out processing according to the difference between the assets and the creditor's right amount, and after docking is completed, the electronic creditor's right voucher is automatically updated or cancelled. The efficiency and safety problems of a traditional mode are solved.

Description

Technical Field

[0001] This invention relates to the field of debt management technology, and in particular to a method for revitalizing and processing the debts of small and micro enterprises based on intelligent matching. Background Technology

[0002] Currently, small and medium-sized enterprises (SMEs) generally face problems such as high accounts receivable ratio and difficulty in capital turnover. Traditional debt revitalization models rely on offline negotiations and manual asset matching, which have pain points such as long process cycles, mismatch between asset and debt value, and cumbersome transfer procedures.

[0003] For example, after transferring receivables, companies need to manually select assets for repayment, which can easily lead to limited asset selection due to information asymmetry; the circulation of receivables certificates relies on paper records, which poses a risk of forgery and is difficult to trace. Therefore, a method for receivables revitalization that combines digital technology is needed to improve the efficiency and security of receivables circulation. Summary of the Invention

[0004] This invention provides a method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching. By building an intelligent platform for receivables transfer, it realizes the full-process digital processing of receivables transfer, asset matching, and certificate transfer, solving the efficiency and security problems of the traditional model.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching includes the following steps: S1, build an intelligent platform for debt transfer, which includes: a debt information storage module, an asset management module, an intelligent matching module, and an electronic voucher management module; S2, Application and verification of debt transfer: SMEs submit applications and related materials. The platform verifies the completeness of the materials and then verifies the authenticity of the debt. Once approved, the debt transferee confirms the application. S3, Tripartite Agreement Signing and Storage: After the assignee of the debt is confirmed, the platform generates the "Debt Transfer Agreement" and completes the signing and sealing of the three parties. After the agreement takes effect, the debt information is stored in the platform for storage to form an unalterable file. S4, Intelligent Asset Matching: Based on the unalterable debt information in the archives, the platform filters assets from multiple dimensions, generates a matching list, and pushes it to the debt transferor; S5, Electronic debt certificate generation: Based on the debt information in the tamper-proof archive, the debt management institution generates an electronic certificate corresponding to the archive, containing a unique blockchain-linked identifier and a redeemable range. The redeemable range corresponds to the matching list, and the certificate is uploaded to the blockchain for evidence storage to ensure authenticity and traceability. S6, Asset Offsetting and Certificate Update: After the debtor selects an asset from the matching list, the platform processes the difference between the asset and the debt amount. After the connection is completed, the electronic debt certificate is automatically updated or cancelled. S7, Debt Certificate Circulation Management: Electronic debt certificates can be circulated among platform users. After the transferor initiates an application and the transferee confirms it, the platform updates the holder information, generates a circulation record, and writes it to the blockchain. The transferee assumes the rights and obligations and can continue to redeem or circulate the certificate.

[0006] Preferably, in the intelligent platform for debt transfer: The debt information storage module is used to store basic debt information, ownership certificates and transfer records, and uses encryption technology to ensure data security. The asset management module is used to collect information on collateral assets provided by the assignee of the claim, including name, specifications, value and ownership status, and to update asset availability in real time. The intelligent matching module is used to automatically match claims and assets based on an algorithm model. The electronic certificate management module is used to generate and manage electronic debt certificates, and supports transfer registration and status updates.

[0007] Preferably, the specific process of S2 is as follows: Small and medium-sized enterprises (SMEs), i.e., debtors, fill out a debt transfer application through the platform and upload the contract signed with the debtor and the accounts receivable confirmation form. After the intelligent platform for debt transfer automatically verifies the completeness of the materials, it connects with the enterprise credit system and contract storage to verify the authenticity of the debt. Once the verification is successful, the debt transferee can confirm the debt.

[0008] Preferably, the specific process of S3 is as follows: After the assignee confirms the assignment, the platform generates an electronic "Debt Transfer Agreement" based on the application information. The agreement is then signed by all three parties through the built-in electronic signature system. Once the agreement takes effect, the debt information, including the debtor, amount, and term, is automatically stored in the platform's evidence storage module, forming an unalterable debt file.

[0009] Preferably, the multi-dimensional asset screening in S4 is based on the following dimensions: Value matching: The deviation between the asset value and the debt transfer price is controlled within ±5%; Industry relevance: Prioritize matching upstream and downstream assets within the industry of the debt transferor; Location preference: Based on the company's registered address, prioritize pushing real estate assets in the surrounding area; Historical preferences: If the enterprise has a history of revitalization, preference labels are generated based on its historical choices; The matching results are sent to the debt transferor in the form of a list, and the asset details can be viewed online.

[0010] Preferably, the intelligent matching module employs a multi-stage hybrid matching algorithm to improve the matching accuracy and efficiency between assets and claims, including five steps: data preprocessing and feature extraction, initial candidate screening, fine ranking, and constraint post-processing. The specific data preprocessing and feature extraction are as follows: Data preprocessing and feature extraction Debt characteristics: Extract the debt amount, maturity date, debtor credit score, contract type, industry classification code, and debt transfer price from the debt file; standardize the numerical characteristics; and perform one-hot coding or embedded coding on the classification characteristics. Asset characteristics: Extract the asset's fair value, ownership status, location, asset type, transfer history, risk score, estimated tax and fees, and actual transferable time window; External features: Access to external data from credit reporting, judicial, real estate registration, and market price indices to calculate the compliance and valuation adjustment factors of debtors and assets; All fields are timestamped upon entry into the database and are stored encrypted with access control.

[0011] Preferably, the specific process for initial candidate screening is as follows: Set a set of assets whose value is within ±X% of the debt amount, whose geographical location is within the radius R of the creditor's registered location, and whose ownership status is transferable or unrestricted. The initial screening results form a candidate set C. Feature vectorization and similarity calculation: For each asset in the candidate set, calculate the relevant matching dimensions, including: Value similarity :

[0012] in, Indicates asset value. Indicates the amount of the claim; Industry relevance : Define the type of relationship between the industry to which the asset belongs and the industry to which the creditor belongs; assign tiered scores according to the closeness of the relationship, with higher scores for closer relationships, specifically divided into three tiers: completely identical industries > direct upstream and downstream relationship > no clear relationship; Geographical distance score :

[0013] Where α is the attenuation parameter, which is based on the platform's historical matching data and fitted with the optimal value through linear regression to ensure that the distance attenuation trend conforms to the actual business scenario. Ownership and Compliance Score : The value is determined based on the clarity of ownership, the absence of seizure records, and the ease of registration, and ranges from 0 to 1. Historical preference score : The matching enhancement score is calculated based on the creditor's historical selection or preference tags; it is also calculated based on the creditor's historical asset selection records, and the frequency percentage of the target asset type in its historical selection is statistically analyzed. The frequency percentage is directly used as the score value, and the higher the frequency percentage, the higher the score.

[0014] The preferred procedure for fine-grained flight ticket sorting and constraint post-processing is as follows: Value similarity Industry relevance Geographical distance score Ownership and Compliance Score and historical preference score The input vector x, representing the j-th asset in the candidate set, is fed into the fine-ranking model to obtain the final score. , The calculation method can be represented by two implementation methods: Lightweight implementation: linear or weighted functions:

[0015] in, The final matching score of the j-th candidate asset is represented by , where i represents the sub-index number, corresponding to the 5 matching dimensions; weight The weight coefficient of the i-th matching dimension represents the contribution of this dimension to the final matching result, which is determined by A / B testing or offline regression training. Enhanced implementation: Supervised learning is performed using a learning ranking model or a deep ranking network; training data comes from historical matching-selection-transaction records as positive samples, and unselected or rejected records as negative samples; Sample weights are introduced during training to correct for differences in sparse categories and values. The specific process of the post-constraint processing is as follows: the candidates after fine ranking are sorted in descending order of score, and a constraint solver is used to ensure that the final candidate assets meet the hard constraints.

[0016] Preferably, in S6, after the debtor selects the asset, the platform automatically calculates: If the asset value equals the debt amount: the debt is directly offset and the asset transfer process is triggered. If the asset value exceeds the debt amount: the company will be prompted to make up the difference online before completing the transfer. If the asset value is lower than the debt amount: the remaining debt is retained in the electronic certificate and can continue to be used to redeem other assets; The platform connects to the real estate registration system. After the transfer registration is completed, the electronic debt certificate will automatically deduct the corresponding amount or be cancelled, and the asset database status will be updated simultaneously.

[0017] Preferably, electronic debt certificates in S7 support circulation among registered members on the platform. During circulation: The transferor initiates a transfer application through the platform, entering the transferee's information and transfer conditions; After the transferee confirms, the platform updates the certificate holder information, generates a transfer record, and writes it into the blockchain; After the transfer is completed, the transferee assumes the rights and obligations of the original holder and continues to exchange assets or transfer them again.

[0018] As can be seen from the above technical solution compared with the prior art, the present invention has the following beneficial effects: 1. This invention combines a fully digital platform with a multi-dimensional intelligent matching algorithm to replace the traditional offline manual operation mode, achieving precise matching of claims and assets, significantly improving the efficiency of claim activation, and alleviating the capital turnover pressure of small and micro enterprises.

[0019] 2. This invention uses blockchain notarization and electronic debt certificate technology to achieve the immutability and traceability of debt information and transfer records, eliminate the risk of certificate forgery, ensure the security and compliance of debt transfer, and expand the channels for the circulation of debt assets. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the steps of the method of the present invention; Figure 2 This is a schematic diagram of the framework for building an intelligent platform for debt transfer in the method of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0022] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention, but should not be used to limit the scope of the present invention.

[0023] like Figure 1 As shown, this invention provides a method for revitalizing and processing the claims of small and micro enterprises based on intelligent matching, including the following steps: S1, build an intelligent platform for debt transfer, which includes: a debt information storage module, an asset management module, an intelligent matching module, and an electronic voucher management module; S2, Application and verification of debt transfer: SMEs submit applications and related materials. The platform verifies the completeness of the materials and then verifies the authenticity of the debt. Once approved, the debt transferee confirms the application. S3, Tripartite Agreement Signing and Storage: After the assignee of the debt is confirmed, the platform generates the "Debt Transfer Agreement" and completes the signing and sealing of the three parties. After the agreement takes effect, the debt information is stored in the platform for storage to form an unalterable file. S4, Intelligent Asset Matching: Based on the unalterable debt information in the archives, the platform filters assets from multiple dimensions, generates a matching list, and pushes it to the debt transferor; S5, Electronic debt certificate generation: Based on the debt information in the tamper-proof archive, the debt management institution generates an electronic certificate corresponding to the archive, containing a unique blockchain-linked identifier and a redeemable range. The redeemable range corresponds to the matching list, and the certificate is uploaded to the blockchain for evidence storage to ensure authenticity and traceability. S6, Asset Offsetting and Certificate Update: After the debtor selects an asset from the matching list, the platform processes the difference between the asset and the debt amount. After the connection is completed, the electronic debt certificate is automatically updated or cancelled. S7, Debt Certificate Circulation Management: Electronic debt certificates can be circulated among platform users. After the transferor initiates an application and the transferee confirms it, the platform updates the holder information, generates a circulation record, and writes it to the blockchain. The transferee assumes the rights and obligations and can continue to redeem or circulate the certificate.

[0024] Example: In this embodiment, a small and medium-sized manufacturing enterprise faces a cash flow problem due to excessively high accounts receivable. The enterprise adopts the method described in this invention to revitalize its receivables: the enterprise submits a receivables transfer application and related materials, which are verified by the platform and the tripartite agreement is signed and stored. The enterprise then obtains a matching asset list pushed by the platform and completes the asset offset with the electronic receivables certificate stored on the blockchain. This ultimately achieves efficient revitalization of receivables and effectively alleviates the enterprise's operating capital pressure.

[0025] First, establish an intelligent platform for debt transfer. For example... Figure 2 As shown, the core modules of the platform include: Debt Information Storage Module: Stores basic debt information, ownership certificates, and transfer records, and uses encryption technology to ensure data security; Asset Management Module: Collects information on collateral assets provided by the assignee of the claim (including name, specifications, value, ownership status, etc.) and updates asset availability in real time; Intelligent matching module: Automatically matches claims and assets based on algorithm models to improve matching accuracy; Electronic certificate management module: Generates and manages electronic debt certificates, and supports transfer registration and status updates.

[0026] Application and verification of debt assignment; Small and medium-sized enterprises (SMEs), acting as debt assignors, fill out a debt assignment application through the platform, uploading materials such as contracts signed with debtors and accounts receivable confirmation slips. After automatically verifying the completeness of the materials, the platform connects with the enterprise credit system and contract notarization platform to verify the authenticity of the debt, such as the validity of the contract and the debtor's confirmation records. Once the verification is passed, the assignment process begins.

[0027] Signing of the tripartite agreement and deposit of debt; After the assignee confirms the assignment, the platform generates an electronic "Debt Transfer Agreement" based on the application information, and completes the signing of all three parties through the built-in electronic signature system. After the agreement takes effect, the target debt information, including the debtor, amount, and term, is automatically stored in the platform's evidence storage module, forming an unalterable debt file.

[0028] Intelligent asset matching; The platform's intelligent matching module filters assets based on the following dimensions: Value matching: The deviation between the asset value and the debt transfer price is controlled within ±5%; Industry relevance: Priority will be given to matching upstream and downstream assets of the industry in which the debt transferor is located, such as production equipment and raw materials for manufacturing enterprises; Location preference: Based on the company's registered address, prioritize pushing real estate assets within a 30-kilometer radius; Historical Preferences: If a company has multiple activation records, preference tags will be generated based on its historical choices. For example, if a company prefers to use commercial properties for mortgage payments, commercial assets will be prioritized.

[0029] The matching results are pushed to the debt transferor in the form of a list, and the asset details, such as ownership certificates and real-life photos, can be viewed online.

[0030] Electronic debt instrument generation; Debt management service providers obtain debt information through the platform and generate electronic debt certificates within 5 business days. Each certificate contains a unique identifier linked to a blockchain address, the debt amount, and the range of assets that can be exchanged for it. Once generated, the certificate is automatically uploaded to the blockchain for evidence storage, ensuring its authenticity and traceability.

[0031] Asset settlement and document update; After the debt assignor selects the asset, the platform automatically calculates: If the asset value equals the debt amount: the debt is directly offset and the asset transfer process is triggered. If the asset value exceeds the debt amount: the company will be prompted to make up the difference online before completing the transfer. If the asset value is lower than the debt amount: the remaining debt is retained in the electronic certificate and can continue to be used to redeem other assets.

[0032] The platform connects to the real estate registration system. After the transfer registration is completed, the electronic debt certificate will automatically deduct the corresponding amount or be cancelled, and the asset database status will be updated simultaneously.

[0033] Management of the transfer of debt instruments; Electronic debt certificates can be transferred among registered members on the platform. During the transfer: The transferor initiates a transfer application through the platform, entering the transferee's information and transfer conditions; Once the transferee confirms, the platform automatically updates the certificate holder information, generates a transfer record, and writes it into the blockchain; After the transfer is completed, the transferee assumes the rights and obligations of the original holder and may continue to exchange assets or transfer them again.

[0034] Taking the revitalization of receivables of a small and micro manufacturing enterprise as an example, the implementation steps are as follows: The company submitted a debt transfer application through the platform, involving RMB 1 million in accounts receivable from debtor Company A, and uploaded the purchase and sale contract and the accounts receivable statement confirmed by Company A; The platform connects to the contract storage system, confirming the authenticity and validity of the contract. Company A has also confirmed the accounts receivable in the system, and the verification is successful. Company B, the assignee of the creditor's rights, confirms the assignment. The platform generates an electronic agreement. After all three parties complete the signing, the creditor's rights information is stored in the blockchain for evidence preservation. The intelligent matching module was activated: The company is registered in XX City, belongs to the auto parts manufacturing industry, and has a historical preference for real estate as collateral. The platform selected 3 industrial plants and 2 shops within a 30-kilometer radius of XX City, valued at 950,000 to 1,050,000 yuan, and pushed them as an asset list. The company selected an industrial plant worth 1.02 million yuan. The platform prompted the company to pay the difference of 20,000 yuan. After online payment, the transfer process was triggered. The debt management service agency generates an electronic debt certificate of 1 million yuan. After the platform completes the transfer registration, the certificate is automatically cancelled, and the status of the factory in the asset database is updated to "paid off". If a company subsequently transfers part of its receivables, such as 500,000 yuan, to supplier Company C, it can initiate a transfer application through the platform. After Company C confirms the application, the certificate holder will be updated to Company C, and the transfer record will be written into the blockchain. Company C can then redeem other assets within the platform using the certificate.

[0035] Example of intelligent matching algorithm: To improve the accuracy and efficiency of matching assets and claims, the intelligent matching module in this invention adopts a multi-stage hybrid matching algorithm, which includes five steps: data preprocessing, feature extraction, initial candidate screening, fine ranking, and constraint post-processing, as detailed below: Data preprocessing and feature construction Debt characteristics: Extract debt amount, maturity date, debtor credit score, contract type, industry classification code, debt transfer price, etc. from the debt file; standardize the numerical characteristics and perform one-hot encoding or embedded encoding on the classification characteristics.

[0036] Asset characteristics: extract the fair value of the asset, ownership status (mortgageable / non-mortgageable), location (latitude and longitude), asset type (real estate / equipment / inventory / accounts receivable), transfer history, mortgage / seizure risk score, estimated tax and fee value, actual transfer window, etc.

[0037] External features: Access to external data such as credit reporting, judicial data, real estate registration data, and market price indices is used to calculate the compliance and valuation adjustment factors for debtors and assets. All fields are timestamped upon entry into the database and are stored encrypted with access control.

[0038] Initial screening of candidates; Asset sets are quickly screened based on simple, computationally inefficient rules. For example, asset values ​​are within ±X% of the debt amount (X=5% by default for broad candidates), and assets are located within a radius R of the creditor's registered address (R=30km by default), with ownership status being transferable / without significant restrictions. The initial screening results form a candidate set C.

[0039] Feature vectorization and similarity calculation: For each asset in the candidate set, calculate the relevant matching dimensions, including: Value similarity :

[0040] in, Indicates asset value. Indicates the amount of the claim; Industry relevance : Define the type of relationship between the industry to which the asset belongs and the industry to which the creditor belongs; assign tiered scores according to the closeness of the relationship, with higher scores for closer relationships, specifically divided into three tiers: completely identical industries > direct upstream and downstream relationship > no clear relationship; Geographical distance score :

[0041] Where α is the attenuation parameter, which is based on the platform's historical matching data and fitted with the optimal value through linear regression to ensure that the distance attenuation trend conforms to the actual business scenario. Ownership and Compliance Score : The value is determined based on the clarity of ownership, the absence of seizure records, and the ease of registration, and ranges from 0 to 1. Historical preference score : The matching enhancement score is calculated based on the creditor's historical selection or preference tags; it is also calculated based on the creditor's historical asset selection records, and the frequency percentage of the target asset type in its historical selection is statistically analyzed. The frequency percentage is directly used as the score value, and the higher the frequency percentage, the higher the score.

[0042] The preferred procedure for fine-grained flight ticket sorting and constraint post-processing is as follows: Value similarity Industry relevance Geographical distance score Ownership and Compliance Score and historical preference score The input vector x, representing the j-th asset in the candidate set, is fed into the fine-ranking model to obtain the final score. , The calculation method can be represented by two implementation methods: Lightweight implementation: linear or weighted functions:

[0043] in, The final matching score of the j-th candidate asset is represented by , where i represents the sub-index number, corresponding to the 5 matching dimensions; weight The weight coefficient of the i-th matching dimension represents the contribution of this dimension to the final matching result, which is determined by A / B testing or offline regression training. Enhanced implementation: Supervised learning is performed using a learning ranking model or a deep ranking network; training data comes from historical matching-selection-transaction records (positive samples) and unselected / rejected records (negative samples). Sample weights are introduced during training to correct for sparse categories and value differences.

[0044] Post-constraint processing involves sorting the ranked candidates in descending order of score and using a constraint solver, rule engine, or integer programming to ensure that the final candidates satisfy hard constraints. For example: If an individual asset or a combination of assets passes the transfer feasibility test, is not subject to seizure, and can be transferred within the stipulated time; Value Coverage Principle: If multiple asset portfolios are chosen for offsetting, the portfolio value must meet the requirements of the debt amount and tax. Risk limit: The type or region of assets that a single creditor or assignee can offset; Hard constraints on user preferences, such as creditors forcibly excluding certain asset classes or geographical regions.

[0045] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0046] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0047] For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media.

[0048] The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0049] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0050] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0051] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0052] The embodiments of the present invention are given for the purposes of illustration and description. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching, characterized in that, Includes the following steps: S1, build an intelligent platform for debt transfer, which includes: a debt information storage module, an asset management module, an intelligent matching module, and an electronic voucher management module; S2, Application and verification of debt transfer: SMEs submit applications and related materials. The platform verifies the completeness of the materials and then verifies the authenticity of the debt. Once approved, the debt transferee confirms the application. S3, Tripartite Agreement Signing and Storage: After the assignee of the debt is confirmed, the platform generates the "Debt Transfer Agreement" and completes the signing and sealing of the three parties. After the agreement takes effect, the debt information is stored in the platform for storage to form an unalterable file. S4, Intelligent Asset Matching: Based on the unalterable debt information in the archives, the platform filters assets from multiple dimensions, generates a matching list, and pushes it to the debt transferor; S5, Electronic debt certificate generation: Based on the debt information in the tamper-proof archive, the debt management institution generates an electronic certificate corresponding to the archive, containing a unique blockchain-linked identifier and a redeemable range. The redeemable range corresponds to the matching list, and the certificate is uploaded to the blockchain for evidence storage to ensure authenticity and traceability. S6, Asset Offsetting and Certificate Update: After the debtor selects an asset from the matching list, the platform processes the difference between the asset and the debt amount. After the connection is completed, the electronic debt certificate is automatically updated or cancelled. S7, Debt Certificate Circulation Management: Electronic debt certificates can be circulated among platform users. After the transferor initiates an application and the transferee confirms it, the platform updates the holder information, generates a circulation record, and writes it to the blockchain. The transferee assumes the rights and obligations and can continue to redeem or circulate the certificate.

2. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 1, characterized in that: In the aforementioned intelligent platform for debt transfer: The debt information storage module is used to store basic debt information, ownership certificates and transfer records, and uses encryption technology to ensure data security. The asset management module is used to collect information on collateral assets provided by the assignee of the claim, including name, specifications, value and ownership status, and to update asset availability in real time. The intelligent matching module is used to automatically match claims and assets based on an algorithm model. The electronic certificate management module is used to generate and manage electronic debt certificates, and supports transfer registration and status updates.

3. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 1, characterized in that: The specific process of S2 is as follows: Small and medium-sized enterprises (SMEs), i.e., debtors, fill out a debt transfer application through the platform and upload the contract signed with the debtor and the accounts receivable confirmation form. After the intelligent platform for debt transfer automatically verifies the completeness of the materials, it connects with the enterprise credit system and contract storage to verify the authenticity of the debt. Once the verification is successful, the debt transferee can confirm the debt.

4. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 1, characterized in that: The specific process of S3 is as follows: After the assignee confirms the assignment, the platform generates an electronic "Debt Transfer Agreement" based on the application information. The agreement is then signed by all three parties through the built-in electronic signature system. Once the agreement takes effect, the debt information, including the debtor, amount, and term, is automatically stored in the platform's evidence storage module, forming an unalterable debt file.

5. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 1, characterized in that: The multi-dimensional asset screening in S4 is based on the following dimensions: Value matching: The deviation between the asset value and the debt transfer price is controlled within ±5%; Industry relevance: Prioritize matching upstream and downstream assets within the industry of the debt transferor; Location preference: Based on the company's registered address, prioritize pushing real estate assets in the surrounding area; Historical preferences: If the enterprise has a history of revitalization, preference labels are generated based on its historical choices; The matching results are sent to the debt transferor in the form of a list, and the asset details can be viewed online.

6. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 5, characterized in that: The intelligent matching module employs a multi-stage hybrid matching algorithm to improve the accuracy and efficiency of matching assets and claims. This includes five steps: data preprocessing and feature extraction, initial candidate screening, fine-grained ranking, and constraint post-processing. The specific details of the data preprocessing and feature extraction are as follows: Data preprocessing and feature extraction Debt characteristics: Extract the debt amount, maturity date, debtor credit score, contract type, industry classification code, and debt transfer price from the debt file; standardize the numerical characteristics; and perform one-hot coding or embedded coding on the classification characteristics. Asset characteristics: Extract the asset's fair value, ownership status, location, asset type, transfer history, risk score, estimated tax and fees, and actual transferable time window; External features: Access to external data from credit reporting, judicial, real estate registration, and market price indices to calculate the compliance and valuation adjustment factors of debtors and assets; All fields are timestamped upon entry into the database and are stored encrypted with access control.

7. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 6, characterized in that: The specific process for the initial screening of candidates is as follows: Set a set of assets whose value is within ±X% of the debt amount, whose geographical location is within the radius R of the creditor's registered location, and whose ownership status is transferable or unrestricted. The initial screening results form a candidate set C. Feature vectorization and similarity calculation: For each asset in the candidate set, calculate the relevant matching dimensions, including: Value similarity : in, Indicates asset value. Indicates the amount of the claim; Industry relevance : Define the type of relationship between the industry to which the asset belongs and the industry to which the creditor belongs; assign tiered scores according to the closeness of the relationship, with higher scores for closer relationships, specifically divided into three tiers: completely identical industries > direct upstream and downstream relationship > no clear relationship; Geographical distance score : Where α is the attenuation parameter, which is based on the platform's historical matching data and fitted with the optimal value through linear regression to ensure that the distance attenuation trend conforms to the actual business scenario; Ownership and Compliance Score : The value is determined based on the clarity of ownership, the absence of seizure records, and the ease of registration, and ranges from 0 to 1. Historical preference score : The matching enhancement score is calculated based on the creditor's historical selection or preference tags; it is also calculated based on the creditor's historical asset selection records, and the frequency percentage of the target asset type in its historical selection is statistically analyzed. The frequency percentage is directly used as the score value, and the higher the frequency percentage, the higher the score.

8. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 7, characterized in that: The specific process for fine-sorting and constraining post-processing of flight tickets is as follows: Value similarity Industry relevance Geographical distance score Ownership and Compliance Score and historical preference score The input vector x, representing the j-th asset in the candidate set, is fed into the fine-ranking model to obtain the final score. , The calculation method can be selected from two implementation methods: Lightweight implementation: linear or weighted functions: in, The final matching score of the j-th candidate asset is represented by , where i represents the sub-index number, corresponding to the 5 matching dimensions; weight The weight coefficient of the i-th matching dimension represents the contribution of this dimension to the final matching result, which is determined by A / B testing or offline regression training. Enhanced implementation: Supervised learning is performed using a learning ranking model or a deep ranking network; training data comes from historical matching-selection-transaction records as positive samples, and unselected or rejected records as negative samples; Sample weights are introduced during training to correct for the difference between sparse classes and value. The specific process of the post-constraint processing is as follows: the candidates after fine ranking are sorted in descending order of score, and a constraint solver is used to ensure that the final candidate assets meet the hard constraints.

9. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 1, characterized in that: After the debtor selects the asset in S6, the platform automatically calculates: If the asset value equals the debt amount: the debt is directly offset and the asset transfer process is triggered. If the asset value exceeds the debt amount: the company will be prompted to make up the difference online before completing the transfer. If the asset value is lower than the debt amount: the remaining debt is retained in the electronic certificate and can continue to be used to redeem other assets; The platform connects to the real estate registration system. After the transfer registration is completed, the electronic debt certificate will automatically deduct the corresponding amount or be cancelled, and the asset database status will be updated simultaneously.

10. The method for revitalizing and processing receivables of small and micro enterprises based on intelligent matching as described in claim 1, characterized in that: The electronic debt certificate in S7 supports circulation among registered members on the platform. During circulation: The transferor initiates a transfer application through the platform, entering the transferee's information and transfer conditions; After the transferee confirms, the platform updates the certificate holder information, generates a transfer record, and writes it into the blockchain; After the transfer is completed, the transferee assumes the rights and obligations of the original holder and continues to exchange assets or transfer them again.

Images