Financing demand matching method and system based on project material depth analysis

Abstract

The invention relates to the technical field of data processing, and particularly discloses a financing demand matching method and system based on project material depth analysis, and the method comprises the steps: obtaining and analyzing project material data of a target financing project, and determining a current period stage of the target financing project and a stage financing demand of the current period stage; obtaining financing resource data of a target financing project, and determining a first candidate subject set, a second candidate subject set, a financing subject sequence, a project candidate subject set and candidate tags of the target financing project; and determining a plurality of policy financing of the target financing project and the project policy label of each policy financing, and optimizing and adjusting the project candidate subject set. The method can accurately determine financing demands and match financing resources, improves financing efficiency, reduces blindness and uncertainty in a financing process, gives consideration to policy bonus and fund safety, reduces financing cost and risk, increases the possibility of financing success, and promotes smooth development of a project.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for matching financing needs based on in-depth analysis of project materials. Background Technology

[0002] Traditional methods of matching financing needs are relatively crude. Early corporate financing relied heavily on limited channels such as bank loans, with business owners estimating their needs based on experience, and communication with financial institutions being relatively simple and direct. However, as the market environment has become increasingly complex and industries have become more segmented, this model has gradually revealed many drawbacks. For example, different industries have vastly different project characteristics. For instance, innovative drug development requires huge upfront investments and has a long cycle, making traditional financing difficult to adapt to its complex needs. Furthermore, construction companies often face slow project cash flow, making traditional financing products unable to meet their continuous funding needs. In recent years, while some new financing models have emerged, such as supply chain financing and equity financing platforms, most are only targeted at single stages or specific types of companies, lacking in-depth analysis of project materials. This results in the accuracy and efficiency of matching financing needs failing to meet the development needs of enterprises. Furthermore, traditional methods do not incorporate natural language processing technologies such as semantic recognition and semantic analysis. They are unable to perform structured analysis and in-depth mining of implicit demand characteristics in market data such as the funding requirements implied in customer cooperation agreements, core related information in technical data such as the equipment investment descriptions required for technology implementation in patent documents, and the semantic logic of policy data clauses. Relying solely on manual screening can easily lead to the omission of key information, resulting in a mismatch between financing needs and resources. This further exacerbates the problem that the accuracy and efficiency of matching financing needs cannot meet the needs of enterprise development.

[0003] Therefore, this invention proposes a financing demand matching method and system based on in-depth analysis of project materials. Summary of the Invention

[0004] This invention provides a method and system for matching financing needs based on in-depth analysis of project materials. By analyzing the acquired project material data, the current cycle stage of the target financing project and its stage-specific financing needs are determined. Combined with acquired financing resource data, a first set of candidate entities, a second set of candidate entities, a sequence of financing entities, a set of project candidate entities, and candidate tags are identified. Based on the project material data, multiple policy-based financing options for the target financing project and the policy tags for each option are determined. The set of project candidate entities for the target financing project is then optimized and adjusted. This method can accurately determine financing needs and match financing resources, improving financing efficiency, reducing blind spots and uncertainties in the financing process, balancing policy benefits with capital security, reducing financing costs and risks, increasing the likelihood of successful financing, providing a more comprehensive and scientific financing solution for the target financing project, and promoting the smooth development of the project.

[0005] This invention provides a method for matching financing needs based on in-depth analysis of project materials, including: S1: Obtain and analyze the project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage; S2: Obtain financing resource data for the target financing project, and based on the current stage of financing needs and financing resource data of the target financing project, determine the first candidate entity set and the second candidate entity set for the target financing project; S3: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, determine the financing entity sequence, project candidate entity set, and candidate tags of the target financing project; S4: Based on the project material data of the target financing project, determine multiple policy financing options for the target financing project and the project policy tags for each policy financing option. Optimize and adjust the set of project candidate entities for the target financing project based on the candidate tags of the target financing project and the project policy tags of all policy financing options.

[0006] Preferably, a financing demand matching method based on in-depth analysis of project materials acquires and analyzes project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage, including: Obtain project material data for the target financing project, including project market data, project technical data, project financial data, project policy data, and project risk data; Obtain the economic industry classification table, analyze the project market data in the project material data, and combine the economic industry classification table to determine the industry to which the target financing project belongs and the full life cycle of the industry; Based on the project material data, the industry to which the target financing project belongs, and the entire life cycle of the industry, determine the current cycle stage of the target financing project and the primary financing need of the current cycle stage. Based on the project material data of the target financing project, its industry, the entire life cycle of the industry, the current cycle stage, the current financing needs of the current cycle stage, and a series of models, the pre-financing needs of the target financing project for each subsequent cycle stage after the current cycle stage of the entire life cycle of the industry are generated. By taking the pre-financing needs of all subsequent cycle stages after the current cycle stage of the target financing project as constraints based on the full life cycle of the industry to which the target financing project belongs, the first financing need of the current cycle stage of the target financing project is optimized and adjusted to determine the stage financing needs of the current cycle stage of the target financing project.

[0007] Preferably, a financing demand matching method based on in-depth analysis of project materials obtains financing resource data for the target financing project, including: Obtain financing resource sub-data for multiple target financing entities of the target financing project. The financing resource sub-data includes financing type, financing amount range, matching requirements data, financing cost data, financing conditions data, and historical financing data. Financing types include debt, equity, and other types. Based on the financing resource sub-data of all target financing entities of the target financing project, the financing resource data of the target financing project is determined.

[0008] Preferably, a financing demand matching method based on in-depth analysis of project materials, based on the current stage of financing demand and financing resource data of the target financing project, determines a first set of candidate entities and a second set of candidate entities for the target financing project, including: Feature extraction is performed on the stage financing needs of the target financing project in the current cycle stage to determine multiple demand features of the target financing project in the current cycle stage and the demand feature value of each demand feature; Feature extraction is performed on the matching requirement data in the financing resource sub-data of each target financing entity of the target financing project to determine multiple matching features for each target financing entity and the matching requirements for each matching feature; Based on each matching feature of each target financing entity, the project material data is extracted to determine the project data for each matching feature of each target financing entity in the target financing project. The matching requirements of each matching feature of each target financing entity are compared with the project data. If the project data of all matching features of the target financing entity meet the matching requirements, the first label of the target financing entity is determined as a candidate entity. If the project data of any matching feature of the target financing entity does not meet the matching requirements, the first label of the target financing entity is determined as a non-candidate entity. Based on all target financing entities whose first label is candidate entity, determine the first candidate entity set for the target financing project; Feature extraction is performed on the financing condition data in the financing resource sub-data of each target financing entity in the first candidate entity set of the target financing project. Multiple condition features of each target financing entity in the first candidate entity set, as well as the condition range and condition label of each condition feature, are determined. The condition label includes performance-based conditions and non-performance-based conditions. Based on the conditional features of each conditional label of each target financing entity in the first candidate entity set that are not performance-based conditions, the project material data is extracted to determine the conditional data of each conditional feature of each target financing entity in the first candidate entity set that are not performance-based conditions. The conditional range and conditional data of each conditional feature of each target financing entity in the first candidate entity set that are not performance-based conditions are compared. If the conditional data of all conditional features of the target financing entity in the first candidate entity set that are not performance-based conditions meet the conditional range, the second label of the target financing entity is determined to be a candidate entity. If the conditional data of any conditional feature of the target financing entity in the first candidate entity set that is not performance-based conditions does not meet the conditional range, the second label of the target financing entity is determined to be a non-candidate entity. Based on the conditional features of each conditional label of each target financing entity in the first candidate entity set as a betting condition, the historical financing data in the financing resource sub-data of each target financing entity in the first candidate entity set is input into the financing prediction model. The predicted conditional value of each conditional label of each target financing entity in the first candidate entity set as a betting condition is determined. It is then determined whether the predicted conditional value of each conditional label of each target financing entity in the first candidate entity set as a betting condition meets the condition range. If the predicted conditional values ​​of all conditional labels of the target financing entity in the first candidate entity set as betting conditions meet the condition range, the third label of the target financing entity is determined as a candidate entity. If the predicted conditional value of any conditional label of the target financing entity in the first candidate entity set as a betting condition does not meet the condition range, the third label of the target financing entity is determined as a non-candidate entity. Based on all target financing entities in the first candidate entity set of the target financing project that are both candidate entities with the second and third labels, determine the second candidate entity set of the target financing project.

[0009] Preferably, a financing demand matching method based on in-depth analysis of project materials determines the financing entity sequence, candidate entity set, and candidate tags for the target financing project based on the second candidate entity set of the target financing project, the stage financing demand of the current cycle stage, and financing resource data, including: The financing type and financing cost data of each target financing entity in the second candidate entity set are input into the cost discount model, and the financing cost value of each target financing entity in the second candidate entity set is determined based on the output of the cost discount model. Based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, and the financing cost value and financing type of each target financing entity in the second candidate entity set of the target financing project, calculate the sequence cost value of each target financing entity in the second candidate entity set of the target financing project. Sort the sequence cost values ​​of all target financing entities in the second candidate entity set of the target financing project from smallest to largest to determine the financing entity sequence of the target financing project; Based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence, calculate the project candidate entity set and candidate labels for the target financing project. Among them, candidate labels include single-backup and multiple-backup.

[0010] Preferably, a financing demand matching method based on in-depth analysis of project materials, based on the project material data of the target financing project, determines multiple policy financing options for the target financing project and the project policy tags for each policy financing option, including: The project policy data in the project materials data is broken down to determine the policy-related financing data of multiple policy financing for the target financing project. The policy-related financing data includes the financing amount range, multiple policy characteristics, the policy characteristic requirements of each policy characteristic, and the stage to which the characteristic belongs. Based on all policy financing data, feature extraction was performed on the project market data, project technology data, project financial data, and project risk data in the project material data to determine multiple project characteristics of the target financing project, feature labels for each project characteristic, and project feature values. Among them, feature labels include market characteristics, technology characteristics, financial characteristics, and risk characteristics. Based on all policy features in the policy-related financing data of each policy financing of the target financing project, extract all project features of the target financing project and determine the policy financing vector of each policy financing of the target financing project; Based on the policy feature requirements, feature stage, and project feature value and feature label of each project feature in the policy financing data of each policy financing of the target financing project, the project policy label of each policy financing of the target financing project is determined. The project policy label includes encouragement and restriction.

[0011] Preferably, a financing demand matching method based on in-depth analysis of project materials optimizes and adjusts the set of candidate project entities for the target financing project based on candidate tags of the target financing project and policy tags of all policy-funded projects, including: If the candidate tag for the target financing project is multiple guarantees and any one of the policy financing projects has an encouragement policy tag, then the set of candidate entities for the target financing project will not be optimized or adjusted, and a letter of intent without prior consent will be signed with each target financing entity in the set of candidate entities for the target financing project. Then, based on the policy label of all projects of the target financing project that are encouraged by the policy, the financing amount range of the target financing project, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence; All projects targeted for financing should apply for policy-based financing that is labeled as "encouraged" under the relevant policy category.

[0012] This invention provides a financing demand matching system based on in-depth analysis of project materials, used to execute any one of the financing demand matching methods based on in-depth analysis of project materials in Examples 1 to 7, including: Analysis module: Acquires and analyzes project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage; The determination module acquires financing resource data for the target financing project, and based on the current stage of financing needs and financing resource data of the target financing project, determines the first set of candidate entities and the second set of candidate entities for the target financing project. Candidate Module: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, determine the financing entity sequence, project candidate entity set, and candidate tags of the target financing project; Adjustment Module: Based on the project material data of the target financing project, determine multiple policy financing options for the target financing project and the project policy tags for each policy financing option. Based on the candidate tags of the target financing project and the project policy tags of all policy financing options, optimize and adjust the set of project candidate entities for the target financing project.

[0013] The beneficial effects of this invention compared to existing technologies are as follows: By analyzing the acquired project material data, the current cycle stage of the target financing project and its stage-specific financing needs are determined. Combined with the acquired financing resource data, a first set of candidate entities, a second set of candidate entities, a financing entity sequence, a project candidate entity set, and candidate tags are identified for the target financing project. Based on the project material data, multiple policy-based financing options for the target financing project and the project policy tags for each policy-based financing option are determined, and the project candidate entity set for the target financing project is optimized and adjusted. This allows for precise identification of financing needs and matching of financing resources, improving financing efficiency, reducing blind spots and uncertainties in the financing process, balancing policy benefits with capital security, reducing financing costs and risks, increasing the likelihood of successful financing, providing a more comprehensive and scientific financing solution for the target financing project, and promoting the smooth development of the project. Moreover, this invention can accurately extract scattered demand features from project materials through natural language processing technologies such as semantic recognition, avoiding the omission of implicit information during manual extraction. It can also deeply analyze the semantic logic of matching requirements and financing conditions in financing resource data, as well as the semantic judgment criteria of "encouraged" and "restricted" categories in policy financing clauses, through natural language processing technologies such as semantic analysis. This further improves the accuracy of financing demand identification and the accuracy of matching financing resources and policies, reducing the mismatch of financing resources caused by human interpretation bias from a technical perspective, strengthening the scientific nature and reliability of financing matching, thereby breaking through the information processing limitations of traditional manual matching and realizing the technical and precise upgrade of the entire financing matching process.

[0014] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0015] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a financing demand matching method based on in-depth analysis of project materials in an embodiment of the present invention; Figure 2 This is a flowchart of a financing demand matching system based on in-depth analysis of project materials, as described in an embodiment of the present invention. Detailed Implementation

[0017] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. Example 1:

[0018] This invention provides a method for matching financing needs based on in-depth analysis of project materials, with reference to... Figure 1 ,include: S1: Obtain and analyze the project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage; S2: Obtain financing resource data for the target financing project, and based on the current stage of financing needs and financing resource data of the target financing project, determine the first candidate entity set and the second candidate entity set for the target financing project; S3: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, determine the financing entity sequence, project candidate entity set, and candidate tags of the target financing project; S4: Based on the project material data of the target financing project, determine multiple policy financing options for the target financing project and the project policy tags for each policy financing option. Optimize and adjust the set of project candidate entities for the target financing project based on the candidate tags of the target financing project and the project policy tags of all policy financing options.

[0019] In this embodiment, project material data for the target financing project is acquired. This data includes information on various aspects such as market, technology, finance, policy, and risks. By analyzing this data, it is determined which stage of the project's life cycle it is currently in, and the stage-specific financing needs it requires at that stage.

[0020] In this embodiment, financing resource data for the target financing project is acquired, including information such as the financing type, financing amount range, and suitability requirements of each target financing entity. This financing resource data is combined with the phased financing needs of the project's current stage. Through analysis of data such as suitability requirements, target financing entities meeting certain conditions are selected, forming a first candidate entity set. Then, further in-depth analysis is conducted on the entities in the first candidate entity set to determine a second candidate entity set.

[0021] In this embodiment, based on the second candidate entity set, combined with the project's current stage of financing needs and financing resource data, the cost value of each candidate entity is calculated, and these cost values ​​are sorted to obtain a financing entity sequence. Then, based on the project's demand characteristics and the financing amount range of each entity in the financing entity sequence, the project candidate entity set and corresponding candidate tags are determined.

[0022] In this embodiment, policy data in the project materials is analyzed to identify multiple policy-based financing options and project policy tags for each policy-based financing option. Then, based on the candidate tags of the project candidate entity set and the project policy tags of the policy-based financing options, the project candidate entity set is optimized and adjusted to make the final candidate entity set more in line with project needs and policy guidance.

[0023] The beneficial effects of the above technologies are as follows: By analyzing the acquired project material data, the current cycle stage of the target financing project and its stage-specific financing needs are determined. Combined with the acquired financing resource data, the first set of candidate entities, the second set of candidate entities, the financing entity sequence, the project candidate entity set, and candidate tags for the target financing project are identified. Based on the project material data, multiple policy-based financing options for the target financing project and the policy tags for each policy-based financing option are determined, and the project candidate entity set for the target financing project is optimized and adjusted. This allows for precise identification of financing needs and matching of financing resources, improving financing efficiency, reducing blind spots and uncertainties in the financing process, balancing policy benefits with capital security, reducing financing costs and risks, increasing the likelihood of successful financing, providing a more comprehensive and scientific financing solution for the target financing project, and promoting the smooth development of the project. Example 2:

[0024] Based on Example 1, a financing demand matching method based on in-depth analysis of project materials is proposed. This method acquires and analyzes project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage, including: Obtain project material data for the target financing project, including project market data, project technical data, project financial data, project policy data, and project risk data; Obtain the economic industry classification table, analyze the project market data in the project material data, and combine the economic industry classification table to determine the industry to which the target financing project belongs and the full life cycle of the industry; Based on the project material data, the industry to which the target financing project belongs, and the entire life cycle of the industry, determine the current cycle stage of the target financing project and the primary financing need of the current cycle stage. Based on the project material data of the target financing project, its industry, the entire life cycle of the industry, the current cycle stage, the current financing needs of the current cycle stage, and a series of models, the pre-financing needs of the target financing project for each subsequent cycle stage after the current cycle stage of the entire life cycle of the industry are generated. By taking the pre-financing needs of all subsequent cycle stages after the current cycle stage of the target financing project as constraints based on the full life cycle of the industry to which the target financing project belongs, the first financing need of the current cycle stage of the target financing project is optimized and adjusted to determine the stage financing needs of the current cycle stage of the target financing project.

[0025] In this embodiment, the project market data focuses on the basic situation of the market in which the project is located, such as the overall scale of the industry (e.g., the annual scale of the electronic waste recycling industry is 50 billion yuan), industry pain points (e.g., the high energy consumption of traditional recycling technology and the low precious metal recycling rate), the target customer group of the project (e.g., new energy vehicle companies and consumer electronics manufacturers), customer penetration rate (e.g., the current customers served by the project account for 5% of the target group), and competitive landscape (e.g., the market share of leading companies is 30%). These data are directly related to the market positioning and growth potential of the project.

[0026] In this embodiment, the project's technical data revolves around the core technology of the project, including the source of the technology (such as independent research and development, industry-university-research cooperation), the maturity of the technology (such as being in the prototype verification, pilot production, or mass production stage), core technical indicators (such as a precious metal recovery rate of 99.95% in chip recycling and a 70% reduction in energy consumption compared to traditional technologies), patent layout (such as having 35 related patents, including 10 invention patents), and the progress of technology industrialization (such as the pilot production line being put into operation with a monthly processing capacity of 100 tons), which are key to judging the project's ability to implement the technology.

[0027] In this embodiment, the project's financial data covers the project's financial status and expectations, such as revenue and profit in the past three years (e.g., revenue of 100 million yuan and net profit of 15 million yuan in 2024), cash flow situation (e.g., monthly cash inflow of 8 million yuan and outflow of 10 million yuan, with a monthly shortfall of 2 million yuan), cost structure (e.g., raw material costs account for 60% and R&D costs account for 15%), and financial forecasts (e.g., expected revenue growth of 50% to 150 million yuan in 2025), which directly reflects the urgency of the project's funding needs and repayment ability. In this embodiment, the project policy data focuses on the alignment between the project and the policy, including whether the project's field complies with national / local industrial policies (such as whether it falls within the scope of electronic waste resource utilization policy support), the policy benefits that can be enjoyed (such as tax reductions, financial subsidies, and special loan interest subsidies), the compliance qualifications already obtained (such as hazardous waste operation licenses and high-tech enterprise certificates), and the policy validity period (such as subsidy policies continuing until 2026), which are related to whether the project can reduce financing costs by leveraging the policy. In this embodiment, project risk data identifies potential risks to the project, such as technological risks (e.g., the progress of research and development of potential alternative technologies, with a 10% probability of biodegradable recycling technology being implemented within 3 years), market risks (e.g., the possibility of a decline in gross profit margin due to intensified industry competition), financial risks (e.g., an overdue accounts receivable rate of 5%, indicating the risk of bad debts), and policy risks (e.g., the cost of equipment modification due to upgrades in environmental standards).

[0028] In this embodiment, the economic industry classification table is a standardized basis for industry division, such as the "Guidelines for Industry Classification of Listed Companies" issued by the China Securities Regulatory Commission, or a detailed industry classification table customized by industry associations. Such tables categorize economic activities into major categories, intermediate categories, and minor categories. For example, electronic waste recycling might be classified under the subcategory of Manufacturing - Waste Resource Comprehensive Utilization - Electronic Waste Recycling and Processing.

[0029] In this embodiment, the industry to which the project belongs is determined by matching the project's market data with the industry description and core business scope in the economic industry classification table. For example, if the core business in the project's market data is the recycling, purification, and resource reuse of waste chips, and the target customers are semiconductor manufacturers and new energy vehicle companies, then by referring to the economic industry classification table, the project's industry can be determined to be the electronic waste recycling and processing industry; if the core business in the project's market data is the research and development and production of new energy vehicle batteries, then it can be matched to the manufacturing industry - automobile manufacturing industry - new energy vehicle manufacturing subcategory.

[0030] In this embodiment, the entire lifecycle is illustrated in various ways. For example, if the project belongs to the new energy vehicle manufacturing industry, the entire lifecycle includes: technology emergence, commercialization, rapid penetration, maturity, and decline. If the project belongs to the integrated circuit (chip design) industry, the entire lifecycle includes: technology verification, productization, scale expansion, red ocean competition, and technology iteration / decline. If the project belongs to the innovative drug (chemical / biological) industry, the entire lifecycle includes: target discovery, preclinical, clinical phase I / II, clinical phase III / NDA, market launch and sales, and patent cliff. If the project belongs to the industrial internet platform industry, the entire lifecycle includes: demand verification, market cultivation, high-speed growth, ecosystem integration, and platform monopoly.

[0031] In this embodiment, the entire industry lifecycle is the macro context, and the current stage of the project's cycle is the micro-position. This needs to be determined by matching the technical, financial, and market data in the project's materials with the characteristics of the industry cycle. For example, if the industry is innovative drugs (chemical drugs), and the project's technical data shows that preclinical research has been completed and Phase I clinical trials are underway; financial data shows that R&D investment is concentrated on clinical trial subject recruitment and data monitoring; and market data shows that no supply agreements have been signed with hospitals, then by comparing the entire lifecycle stages of target discovery, preclinical, Phase I / II clinical trials, etc., the current cycle stage is determined to be Phase I / II clinical trials. If the industry is an industrial internet platform, and the project's market data shows that 20 small and medium-sized manufacturing plants have been signed, the platform's monthly active customers (using third-party tags to identify entities with high feasibility for betting) are growing at 50%; technical data shows that core functions (equipment networking, data dashboards) are operating stably and a supply chain collaboration module is under development; and financial data shows that revenue has increased by 120% year-on-year, still relying on financing to cover R&D, then by comparing the entire lifecycle stages of demand verification, market cultivation, high-speed growth, etc., the current cycle stage is determined to be a high-speed growth period.

[0032] In this embodiment, the first financing requirement is a preliminary funding requirement derived from the core objectives of the current stage, which needs to be determined in conjunction with the funding gap and priority of use in the project materials data. For example: In the Phase I / II clinical trial stage of innovative drugs (chemical drugs), the core objective is to complete the Phase I / II clinical trial and submit the trial data to the drug regulatory authority. Financial data shows that 80 million yuan is needed for clinical trials and 5 million yuan for patent maintenance in the next 18 months. Current cash can only cover 30 million yuan, so the first financing requirement is 55 million yuan. The priority for use is clinical trials (80 million - 30 million = 50 million) + patent maintenance (5 million). Equity dilution (not exceeding 25% in a single round) is acceptable. Priority should be given to institutions with investment experience in the pharmaceutical industry. In the high-growth stage of industrial internet platforms, the core objective is to expand customer coverage (from 20 to 50) and develop supply chain collaboration modules. Financial data shows that customer expansion requires 30 million yuan for marketing expenses and 20 million yuan for module R&D. The cash flow gap is 40 million yuan, so the first financing requirement is 40 million yuan. The priority for use is marketing (30 million) + R&D (10 million). Debt financing with an annualized rate of less than 8% or equity dilution with a rate of less than 15% is acceptable.

[0033] In this embodiment, based on the current status of the project, the entire industry lifecycle, and a series of models (the model logic is industry patterns + project characteristics + historical data), the financing needs of all subsequent stages of the cycle are predicted in a forward-looking manner, avoiding a disconnect between short-term financing and long-term development. The scope of the subsequent stages is clearly defined: the subsequent stages are all stages following the current stage. For example, for innovative drugs (chemical drugs) currently in Phase I / II clinical trials, the subsequent stages are Phase III clinical trials / NDA → market launch and mass production → patent cliff; for industrial internet platforms currently in a period of rapid growth, the subsequent stages are ecosystem integration → platform monopoly.

[0034] In this embodiment, the financing scale, purpose, and cost tolerance range for each post-cycle stage are determined by combining the core objectives of each stage of the industry's entire life cycle, project material data (such as technology iteration plans and market expansion plans), and a series of models (referencing the financing needs patterns of similar projects in the same industry at the corresponding stages). For example: Phase III clinical trials / NDA stage (post-cycle stage 1) for innovative drugs (chemical drugs): The core objective is to complete large-scale clinical trials and submit an NDA (New Drug Application). Referring to industry data (average investment in Phase III clinical trials is 120 million RMB), and considering the large scale of the project (covering 50 hospitals), the pre-financing requirement is 150 million RMB, to be used for Phase III subject recruitment (120 million RMB) + NDA application fees (30 million RMB). Equity dilution of up to 20% is acceptable, and investors must have resources to communicate with the drug regulatory authority. Industrial Internet platform ecosystem integration stage (post-cycle stage 1): The core objective is to connect 10 upstream and downstream service providers (such as logistics and raw material suppliers) and develop cross-enterprise collaborative functions. Referring to industry data (average investment in the ecosystem integration stage is 80 million RMB), and considering the large existing customer base (50 factories), the pre-financing requirement is 100 million RMB, to be used for service provider access subsidies (60 million RMB) + function development (40 million RMB). Priority will be given to industrial capital (such as leading manufacturing enterprises), and annualized debt financing of up to 6% is acceptable.

[0035] In this embodiment, the pre-financing needs of the later-stage phase are used as constraints to adjust the first financing needs in reverse, ensuring that current financing not only meets short-term needs but also paves the way for future phases, avoiding short-term decisions from impacting long-term development. The core logic of the constraints is that the constraints of the pre-financing needs of the later-stage phase are mainly reflected in three aspects: first, the connection of financing costs (current financing costs are too high, and future debt replacement pressure will be high); second, equity dilution control (excessive dilution at present will lead to loss of control in the future); and third, the compatibility of fund usage (current funds only take into account short-term turnover, delaying future technology layout). Optimization and Adjustment Process and Results: For example, the initial financing need during the high-growth phase of an industrial internet platform is 40 million yuan, including 30 million yuan in equity financing (15% dilution) and 10 million yuan in debt financing (10% annualized return), to be used for marketing (30 million yuan) and R&D (10 million yuan). The pre-financing needs in the later-stage ecosystem integration phase impose constraints: future industrial capital needs to be introduced, and the current 15% equity dilution may result in the founder's control falling below 50% when industrial capital invests in the future; the current 10% annualized debt is too high, leading to high future replacement costs. Based on these constraints, the optimization and adjustments are as follows: First, reduce equity dilution by adjusting the 30 million yuan equity financing (15% dilution) to 20 million yuan equity financing (10% dilution); second, reduce the debt cost... First, the 10 million yuan ordinary loan with an annualized interest rate of 10% will be replaced with a 15 million yuan policy-subsidized loan with an annualized interest rate of 6% (based on the 'Industrial Internet Special Subsidy' qualification in the project policy data); second, the use of funds will be adjusted, with 5 million yuan taken from the 30 million yuan marketing budget for preliminary research on ecosystem integration (laying the groundwork for the later cycle stage); the final determined financing needs for the current cycle stage are 35 million yuan, including 20 million yuan equity financing (diluted by 10%, prioritizing VCs with manufacturing resources) + 15 million yuan policy-subsidized loan (6% annualized interest rate), to be used for marketing (25 million yuan) + R&D (5 million yuan) + ecosystem research (5 million yuan), which not only meets the current high-speed growth needs, but also reserves equity and cost space for the later cycle ecosystem integration.

[0036] The beneficial effects of the above technologies are as follows: acquiring and analyzing project material data of the target financing project, determining the current cycle stage of the target financing project and the stage financing needs of the current cycle stage, can improve the accuracy of financing demand matching, reduce the risk of resource mismatch, avoid the disconnect between short-term financing and long-term development, and achieve the continuity of full-cycle capital planning. Example 3:

[0037] Based on Example 1, the financing resource data of the target financing project is obtained, including: Obtain financing resource sub-data for multiple target financing entities of the target financing project. The financing resource sub-data includes financing type, financing amount range, matching requirements data, financing cost data, financing conditions data, and historical financing data. Financing types include debt, equity, and other types. Based on the financing resource sub-data of all target financing entities of the target financing project, the financing resource data of the target financing project is determined.

[0038] In this embodiment, detailed information on the financing resources that the target financing entity (i.e., various institutions or entities that may provide funds for the project) can provide is collected.

[0039] In this embodiment, the target financing entity is a potential suitable entity initially screened based on the preliminary analysis of the project (such as the industry, current cycle stage, and financing needs).

[0040] In this embodiment, the financing type clearly defines the category of the financing provided by the entity: bond type: such as working capital loans provided by banks, corporate bonds issued by enterprises, and interest-subsidized loans from policy banks; equity type: such as equity investment provided by VC / PE, seed round investment by angel investors, and strategic investment by industrial capital; other type: such as equipment leasing financing by financial leasing companies, accounts receivable financing by supply chain finance platforms, and fiscal subsidies provided by the government (not direct loans / equity, classified as other).

[0041] In this embodiment, the financing amount range is the range of funds that each entity can provide, and it needs to be accurate to a specific value. For example, the financing amount range for AVC (equity type) is 20 million to 50 million yuan; the working capital loan amount range for Bank B (bond type) is 5 million to 200 million yuan; and the subsidy amount range for Industry Fund C (other type) is 1 million to 10 million yuan.

[0042] In this embodiment, the data on the adaptation requirements serves as the entry threshold for financing projects and is the key to determining whether a project can be matched with the resource. Industry adaptation: for example, AVC institutions only adapt to projects in the hard technology and environmental protection industries, and Bank B only supports companies with revenue exceeding 100 million yuan and no overdue records. Qualification adaptation: for example, C industry fund requires projects to have high-tech enterprise qualifications.

[0043] In this embodiment, the financing cost data represents the price a project must pay to obtain the financing. Different financing types have different cost representations. For bonds, the annualized interest rate is the primary indicator, such as the annualized interest rate of Bank B's loan: LPR + 80BP (current LPR is 3.45%, actual annualized rate is 4.25%), and some also include handling fees (such as 0.5% of the loan amount). For equity, the equity dilution ratio or performance-based clauses are used, such as AVC investing 50 million yuan, requiring 10% equity dilution, and attaching a performance-based clause requiring revenue to exceed 500 million yuan within 3 years. Other types have large cost differences, such as the rental interest rate of financial leasing (5% annualized) and government subsidies (0 cost, but subject to fund usage requirements).

[0044] In this embodiment, the financing conditions data are the constraints proposed by the entity other than costs, which directly affect the feasibility of financing. For bonds: guarantee requirements (e.g., Bank B requires real estate mortgage with a loan-to-value ratio of 60%), repayment methods (e.g., quarterly interest payments and principal repayment at maturity), and restrictions on the use of funds (e.g., only for equipment procurement); For equity: post-investment management requirements (e.g., AVC requires the appointment of one director to participate in major decisions) and exit requirements (e.g., exit through IPO or M&A within 5 years); Other types: such as supply chain finance requiring confirmation of ownership by the core enterprise (e.g., if the downstream customer of the project is a large car manufacturer, the car manufacturer needs to confirm accounts receivable), and government subsidies requiring quarterly reports on the use of funds.

[0045] In this embodiment, historical financing data refers to the main body's past financing cases. For example, AVC's historical financing data shows that it has invested in 10 environmental protection projects in the past 3 years, of which 8 projects completed the next round of financing during the growth stage, indicating that it has strong resource integration capabilities for growth-stage projects in the environmental protection industry; Bank B's historical data shows that it has provided loans to 5 electronic waste recycling companies in the past year with no overdue records, indicating that it has sufficient experience in risk control in this industry.

[0046] The beneficial effects of the above technologies are: obtaining financing resource data for the target financing project can provide data support for determining the first and second candidate sets of the target financing project. Example 4:

[0047] Based on Example 2, and using the current stage of financing needs and financing resource data of the target financing project, a first set of candidate entities and a second set of candidate entities for the target financing project are determined, including: Feature extraction is performed on the stage financing needs of the target financing project in the current cycle stage to determine multiple demand features of the target financing project in the current cycle stage and the demand feature value of each demand feature; Feature extraction is performed on the matching requirement data in the financing resource sub-data of each target financing entity of the target financing project to determine multiple matching features for each target financing entity and the matching requirements for each matching feature; Based on each matching feature of each target financing entity, the project material data is extracted to determine the project data for each matching feature of each target financing entity in the target financing project. The matching requirements of each matching feature of each target financing entity are compared with the project data. If the project data of all matching features of the target financing entity meet the matching requirements, the first label of the target financing entity is determined as a candidate entity. If the project data of any matching feature of the target financing entity does not meet the matching requirements, the first label of the target financing entity is determined as a non-candidate entity. Based on all target financing entities whose first label is candidate entity, determine the first candidate entity set for the target financing project; Feature extraction is performed on the financing condition data in the financing resource sub-data of each target financing entity in the first candidate entity set of the target financing project. Multiple condition features of each target financing entity in the first candidate entity set, as well as the condition range and condition label of each condition feature, are determined. The condition label includes performance-based conditions and non-performance-based conditions. Based on the conditional features of each conditional label of each target financing entity in the first candidate entity set that are not performance-based conditions, the project material data is extracted to determine the conditional data of each conditional feature of each target financing entity in the first candidate entity set that are not performance-based conditions. The conditional range and conditional data of each conditional feature of each target financing entity in the first candidate entity set that are not performance-based conditions are compared. If the conditional data of all conditional features of the target financing entity in the first candidate entity set that are not performance-based conditions meet the conditional range, the second label of the target financing entity is determined to be a candidate entity. If the conditional data of any conditional feature of the target financing entity in the first candidate entity set that is not performance-based conditions does not meet the conditional range, the second label of the target financing entity is determined to be a non-candidate entity. Based on the conditional features of each conditional label of each target financing entity in the first candidate entity set as a betting condition, the historical financing data in the financing resource sub-data of each target financing entity in the first candidate entity set is input into the financing prediction model. The predicted conditional value of each conditional label of each target financing entity in the first candidate entity set as a betting condition is determined. It is then determined whether the predicted conditional value of each conditional label of each target financing entity in the first candidate entity set as a betting condition meets the condition range. If the predicted conditional values ​​of all conditional labels of the target financing entity in the first candidate entity set as betting conditions meet the condition range, the third label of the target financing entity is determined as a candidate entity. If the predicted conditional value of any conditional label of the target financing entity in the first candidate entity set as a betting condition does not meet the condition range, the third label of the target financing entity is determined as a non-candidate entity. Based on all target financing entities in the first candidate entity set of the target financing project that are both candidate entities with the second and third labels, determine the second candidate entity set of the target financing project.

[0048] In this embodiment, the phased financing needs of the project's current cycle stage are broken down into comparable demand characteristics and assigned specific characteristic values. These characteristics must cover the core dimensions of the phased financing needs. For example, if the phased financing need is 15 million yuan, used for core equipment upgrades (8 million yuan), market expansion (5 million yuan), and cash flow reserves (2 million yuan), with acceptable equity dilution ≤12% and annualized debt interest rate ≤8%, and the investor must possess semiconductor industry chain resources, then the extracted demand characteristics include financing amount, use of funds, upper limit of equity dilution, upper limit of debt interest rate, and type of investor resources. Determination of demand characteristic values: Characteristic values ​​are specific quantitative or qualitative descriptions of the demand characteristics and must be clearly defined based on project data. For example, the characteristic values ​​corresponding to the above demand characteristics are: financing amount = 15 million yuan, use of funds = equipment upgrades + market expansion + cash flow reserves, upper limit of equity dilution = 12%, upper limit of debt interest rate = 8%, and type of investor resources = semiconductor industry chain resources. These characteristic values ​​serve as the benchmark for subsequently determining whether the financing entity is suitable.

[0049] In this embodiment, by comparing the matching characteristics of the financing entity with the demand characteristics of the project, it is initially determined whether the entity meets the basic access requirements, and the first label is used to distinguish between candidates and non-candidates.

[0050] In this embodiment, the matching features are a breakdown of the matching requirements data of the financing entity, reflecting the entity's basic entry threshold for the project. For example, the matching requirements data of a certain VC institution (target financing entity) are: focusing on the semiconductor and environmental protection industry, investing in growth-stage projects, with a single investment of 20 million to 50 million yuan, and accepting equity dilution of ≤15%. Then the extracted matching features include the matching industry, the matching project cycle, the range of single investment amount, and the acceptance of equity dilution. The corresponding matching requirements (i.e., the qualified range of features) are: matching industry = semiconductor and environmental protection, matching project cycle = growth stage, single investment amount range = 20 million to 50 million yuan, and acceptance of equity dilution ≥15%.

[0051] In this embodiment, project data corresponding to the matching characteristics is extracted from the project material data. For example, for the matching characteristics of the above-mentioned VC institutions, the extracted project data is: Project Industry = Semiconductor Environmental Protection (Market Data), Stage Financing Requirement Amount = RMB 15 million (Financial Data), Acceptable Equity Dilution Limit = 12% (Requirement Characteristic Value).

[0052] In this embodiment, if all project data with matching characteristics meet the matching requirements, the label is a candidate entity; if any one does not meet the requirements, it is a non-candidate entity. For example, if the single investment amount range of the aforementioned VC institution is 20 million to 50 million yuan, but the project demand amount is only 15 million yuan (not met), then the first label of the VC institution is a non-candidate entity; if the matching requirements of a bank are matching industry = environmental protection, single loan of 5 million to 20 million yuan, and annualized interest rate ≤ 8%, and the project data meets all the requirements (industry = environmental protection, demand amount of 15 million yuan within 5 million to 20 million yuan, acceptable interest rate ≤ 8%), then the first label of the bank is a candidate entity.

[0053] In this embodiment, the financing conditions data of the first set of candidate entities that have passed the initial screening (all financing entities whose first label is candidate entity) are further broken down to clarify the condition characteristics, condition range and condition labels (betting / non-betting) in preparation for the second screening.

[0054] In this embodiment, the conditional features are the specific constraints proposed by the financing entity, and the conditional range is the qualified range of the constraints. For example, the financing conditional data of a first-choice entity (PE institution) is that it will appoint one director after investment, and the funds must be reported quarterly. The performance-based clause is: cumulative revenue ≥ 800 million yuan from 2025 to 2027, and annualized return rate not less than 15%. Then the extracted conditional features include post-investment management requirements, fund usage reporting requirements, performance-based revenue target, and performance-based return rate target. The corresponding conditional range is: post-investment management requirement = appoint one director, fund usage reporting requirement = quarterly reporting, performance-based revenue target ≥ 800 million yuan (2025-2027), and performance-based return rate target ≥ 15%.

[0055] In this embodiment, non-performance-based conditions are constraints without performance commitments, and are usually immediately verifiable or routine management requirements, such as post-investment management requirements and reporting requirements for the use of funds. Performance-based conditions are constraints that require the project to achieve future performance targets, such as performance-based revenue targets and performance-based rate of return targets. These conditions require prediction of whether the project can achieve them in the future, so they are tagged separately.

[0056] In this embodiment, it is determined whether the non-betting conditions of the first candidate subject match the current capabilities of the project, and the second label is used to further narrow down the candidate range.

[0057] In this embodiment, conditional data corresponding to the non-performance-based conditions is extracted from the project material data, i.e., whether the project currently has the ability to meet the condition. For example, for the post-investment management requirement = appointing 1 director, the extracted project data is that the project company's articles of association allow the investor to appoint 1 director (basic data); for the fund usage reporting requirement = quarterly reporting, the extracted project data is that the project's existing financial team has the ability to report quarterly (operational data).

[0058] In this embodiment, if all non-betting condition data meet the condition range, the entity is labeled as a candidate entity; if any one condition is not met, the entity is not a candidate entity. For example, if a first candidate entity (bank) requires factory mortgage with a loan-to-value ratio of ≤60%, and project data shows that its own factory is valued at 100 million yuan and can be mortgaged (a loan-to-value ratio of 60% is 60 million yuan, which meets the loan requirements), then the second label is a candidate entity. If an entity requires the project to have more than 3 years of operation, but the project has only been operating for 2 years (not met), then the second label is a non-candidate entity.

[0059] In this embodiment, a predictive model is used to assess whether a project can meet the performance conditions in the future, avoiding the limitations of judging solely based on current data, and a third label is used to screen entities with high feasibility for performance.

[0060] In this embodiment, the historical financing data of the first candidate entity is input into the financing prediction model. For example, the historical financing data of a certain PE institution is that it has invested in 5 growth-stage semiconductor environmental protection projects in the past 3 years. Among them, 4 projects have achieved the revenue target of the performance guarantee (average completion rate of 110%), and the average revenue growth rate of the projects is 45%. These data can help the model judge the rationality of the performance guarantee terms of the institution and the achievement pattern of similar projects.

[0061] In this embodiment, the financing forecasting model combines historical financing data, project material data (such as current revenue growth rate, market expansion plans), and industry trends to output a predicted value for the project's future performance targets. For example, for a PE institution's performance target of ≥800 million RMB in cumulative revenue from 2025 to 2027, the model predicts the project's cumulative revenue over the next three years to be 850 million RMB (predicted value).

[0062] In this embodiment, if the predicted condition values ​​of all the betting condition features meet the condition range, the label is a candidate entity; if any one does not meet the condition, it is a non-candidate entity. For example, if the predicted condition value of 850 million yuan ≥ 800 million yuan (meets the condition), then the third label is a candidate entity; if the model predicts that the cumulative revenue is only 750 million yuan (does not meet the condition), then the third label is a non-candidate entity.

[0063] In this embodiment, the second and third labels are combined to filter out entities that simultaneously meet both the non-betting and betting conditions, forming the final precise candidate pool.

[0064] In this embodiment, only financing entities with both the second and third labels as candidate entities are retained from the first candidate entity set. For example, if a PE institution has the first label as a candidate entity (meeting the suitability requirements), the second label as a candidate entity (meeting the non-performance-based conditions: appointed directors + capital reporting), and the third label as a candidate entity (meeting the performance-based condition of achieving projected revenue targets), it is included in the second candidate entity set. If a bank has the second label as a candidate entity (meeting the collateral requirements) but no performance-based conditions (the third label is assumed to meet the conditions), it is also included in the second candidate entity set. The final second candidate entity set is a pool of entities highly adapted to the financing needs of the project stage, providing precise resources for matching subsequent financing solutions.

[0065] The beneficial effects of the above technologies are as follows: Based on the current stage of financing needs and financing resource data of the target financing project, the first and second candidate sets of the target financing project can be determined. This can avoid the one-sidedness of single-dimensional screening, improve the foresight of the feasibility assessment of the performance guarantee, accurately identify multiple candidate entities that meet both the basic requirements and the conditional constraints, and reduce the financing matching risk. Example 5:

[0066] Based on Example 4, and using the second candidate entity set of the target financing project, the phased financing needs of the current cycle stage, and financing resource data, the financing entity sequence, candidate entity set, and candidate tags of the target financing project are determined, including: The financing type and financing cost data of each target financing entity in the second candidate entity set are input into the cost discount model, and the financing cost value of each target financing entity in the second candidate entity set is determined based on the output of the cost discount model. Based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, and the financing cost value and financing type of each target financing entity in the second candidate entity set of the target financing project, calculate the sequence cost value of each target financing entity in the second candidate entity set of the target financing project. Sort the sequence cost values ​​of all target financing entities in the second candidate entity set of the target financing project from smallest to largest to determine the financing entity sequence of the target financing project; Based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence, calculate the project candidate entity set and candidate labels for the target financing project. Among them, candidate labels include single-backup and multiple-backup.

[0067] In this embodiment, it is clearly defined whether the target financing entity provides debt, equity, or other types of financing, which forms the basis for the model's selection of conversion logic. For example, a bank provides debt loans, while a VC firm provides equity investments. Since their cost measurement methods differ, different conversion rules are required. The model uses industry benchmark costs, time value of money, risk premiums, and other dimensions to convert the original costs of different types into a unified financing cost value (usually in percentage form for easy horizontal comparison). For example: For bond financing: Bank loans have an annualized rate of 4.25% + handling fees of 0.5%. Considering a 2-year funding period, the financing cost is 4.5% (handling fees are spread over 2 years, increasing by 0.25% annually, totaling 4.25% + 0.25% = 4.5%). For equity financing: VC investment dilutes 10% equity. Based on a projected valuation of 1 billion yuan over the next 3 years, 10% equity corresponds to a value of 100 million yuan. If the current investment is 50 million yuan, combined with the risk premium for failure (calculated at a 10% probability, compensation of 5 million yuan is equivalent to a current cost of 500,000 yuan), the financing cost is 8% (100 million yuan implicit cost + 500,000 yuan risk cost) ÷ (50 million yuan × 3 years) ≈ 8%). Ultimately, a unified financing cost value is determined for each target financing entity, such as 4.5% for banks, 8% for VC, and 5% for financial leasing, achieving comparability of financing costs for different types of financing.

[0068] In this embodiment, based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, and the financing cost value and financing type of each target financing entity in the second candidate entity set of the target financing project, the sequence cost value of each target financing entity in the second candidate entity set of the target financing project is calculated. The formula for calculating the sequence cost value can be expressed as: ; in, Let represent the sequence cost value of the a-th target financing entity in the second set of candidate entities for the target financing project. Let represent the financing cost value of the a-th target financing entity in the second set of candidate entities for the target financing project. This represents the b-th demand characteristic in the current cycle stage of the target financing project. This represents the demand characteristic value of the b-th demand characteristic in the current cycle stage of the target financing project. This represents the financing type in the financing resource sub-data of the a-th target financing entity in the second candidate entity set of the target financing project. Na represents the directional factor of the a-th target financing entity in the second candidate entity set of the target financing project, and N2 represents the number of demand characteristics in the current cycle stage of the target financing project.

[0069] In this embodiment, This indicates that when the b-th demand characteristic of the target financing project in its current cycle stage is equity-oriented, the value is taken as [value]. , The value ranges from 0 to 1. The closer the value is to 1, the more the target financing project is inclined to choose equity financing. The closer the value is to 0, the less the target financing project is inclined to choose equity financing.

[0070] In this embodiment, based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence, the project candidate entity set and candidate labels of the target financing project are calculated. The calculation formula for the project candidate entity set and candidate labels can be expressed as follows: ; Wherein, SC represents the set of candidate entities for the target financing project, and CA represents the candidate tags for the target financing project. These represent the 1st, cth, and Nuth target financing entities in the financing entity sequence, respectively. Indicates the pad factor, This represents the sequence cost value of the c-th target financing entity in the sequence of financing entities, 1 c N2, Nu represents the number of target financing entities among the project candidate entities.

[0071] In this embodiment, This indicates that the value is taken as the b-th demand characteristic in the current cycle stage of the target financing project when the demand amount is the value. .

[0072] In this embodiment, This indicates the total amount of demand for the target financing project based on the demand amount and the safety cushion factor.

[0073] In this embodiment, the safety pad factor The value can be 0.1.

[0074] In this embodiment, the number Nu of target financing entities among the project candidate entities represents the number of project candidate entities accumulated when the accumulated sequence cost value is greater than the total demand amount value.

[0075] The beneficial effects of the above technologies are as follows: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, the financing entity sequence, project candidate entity set, and candidate tags of the target financing project can be determined. This can objectively assess the cost of the target financing entity, avoid a single cost orientation, balance cost optimization and capital coverage security, and reduce the risk of financing interruption. Example 6:

[0076] Based on Example 1, and using the project material data of the target financing project, multiple policy financings for the target financing project and the project policy tags for each policy financing are determined, including: The project policy data in the project materials data is broken down to determine the policy-related financing data of multiple policy financing for the target financing project. The policy-related financing data includes the financing amount range, multiple policy characteristics, the policy characteristic requirements of each policy characteristic, and the stage to which the characteristic belongs. Based on all policy financing data, feature extraction was performed on the project market data, project technology data, project financial data, and project risk data in the project material data to determine multiple project characteristics of the target financing project, feature labels for each project characteristic, and project feature values. Among them, feature labels include market characteristics, technology characteristics, financial characteristics, and risk characteristics. Based on all policy features in the policy-related financing data of each policy financing of the target financing project, extract all project features of the target financing project and determine the policy financing vector of each policy financing of the target financing project; Based on the policy feature requirements, feature stage, and project feature value and feature label of each project feature in the policy financing data of each policy financing of the target financing project, the project policy label of each policy financing of the target financing project is determined. The project policy label includes encouragement and restriction.

[0077] In this embodiment, policy financing that can be accessed is extracted from project policy data, and the core information of each policy is clarified. The scope of project policy data includes national / local industrial policies, subsidy documents, tax incentive policies, and special fund management methods related to the field to which the project belongs, such as the national electronic waste resource utilization subsidy policy and the local semiconductor industry special fund management method, and the tax reduction and exemption policy for environmental protection enterprises.

[0078] In this embodiment, the financing amount range indicates the range of funds that the policy can provide, such as the range of 2 million to 5 million yuan for electronic waste resource utilization subsidies, and the range of 10 million to 30 million yuan for semiconductor industry special funds.

[0079] In this embodiment, multiple policy features represent the core requirements of the policy for the project. For example, the policy features for subsidies for the recycling of electronic waste include annual processing volume, technical and environmental standards, and R&D investment ratio; the policy features for tax reduction and exemption policies include the company's registered location, revenue scale, and environmental qualifications. Each policy feature's requirements specify the eligibility criteria, such as an annual processing volume ≥ 1000 tons, technical and environmental standards meeting national level one, R&D investment ratio ≥ 5%, and the company's registered location being in a local high-tech zone. The stage to which a feature belongs corresponds to the project cycle stage, i.e., which stage of the project the requirement applies to.

[0080] In this embodiment, project features related to policy characteristics are extracted from multi-dimensional project data, and their type labels and specific values ​​are clearly defined, enabling a comparable correspondence between project features and policy characteristics. Based on the identified policy characteristics of all policy financing, corresponding project features are extracted from four categories of project material data to ensure coverage of all dimensions of policy concern: Market features related to policy are extracted from project market data, such as the percentage of new energy vehicle companies served, labeled as a market feature, based on the policy characteristic of target customer type; Technical features related to policy are extracted from project technical data, such as pollutant emission concentration, labeled as a technical feature, based on the policy characteristic of environmental protection technical standards; Financial features related to policy are extracted from project financial data, such as the 2024 R&D expense / revenue ratio, labeled as a financial feature, based on the policy characteristic of R&D investment ratio; Risk features related to policy are extracted from project risk data, such as the number of environmental violations in the past three years, labeled as a risk feature, based on the policy characteristic of compliance risk level.

[0081] In this embodiment, based on the policy feature requirements, feature stage, and project feature value and feature label of each project feature in the policy-type financing data of each policy financing of the target financing project, the project policy label of each policy financing of the target financing project is determined. The formula for calculating the project policy label can be expressed as follows: ; in, This indicates the project policy label for the i-th policy financing of the target financing project. Th1 represents the project matching value of the i-th policy financing of the target financing project, and Th1 represents the first matching threshold. The feature label represents the j-th project feature in the policy financing vector of the i-th policy financing of the target financing project. This represents the k-th policy feature of the i-th policy financing of the target financing project. This represents the j-th project feature in the policy financing vector of the i-th policy financing of the target financing project. This represents the project feature value of the j-th project feature in the policy financing vector of the i-th policy financing of the target financing project. Let CP represent the policy feature requirement of the k-th policy feature of the i-th policy financing of the target financing project, and let CP represent the current cycle stage of the target financing project. Let i represent the stage to which the k-th policy feature of the i-th policy financing of the target financing project belongs, and let iN1 represent the number of project features in the policy financing vector of the i-th policy financing of the target financing project. The project feature value of the j-th project feature in the policy financing vector of the i-th policy financing of the target financing project, and the sub-matching value based on the policy feature requirement of the k-th policy feature.

[0082] In this embodiment, the project matching value This represents the project feature value of the j-th project feature in the policy financing vector of all policy financing projects for the target financing project, and the project matching value based on the policy feature requirements of the corresponding policy feature.

[0083] In this embodiment, the first matching threshold Th1 can be 0.6.

[0084] The beneficial effects of the above technologies are as follows: Based on the project material data of the target financing project, multiple policy financing options for the target financing project and the project policy tags for each policy financing option can be determined. This can accurately match the compatibility between the project and the policy, realize a visual assessment of policy compatibility, provide clear policy basis for financing decisions, and improve the efficiency of policy dividend utilization. Example 7:

[0085] Based on Example 6, the set of candidate entities for the target financing project is optimized and adjusted according to the candidate tags of the target financing project and the policy tags of all policy-funded projects, including: If the candidate tag for the target financing project is multiple guarantees and any one of the policy financing projects has an encouragement policy tag, then the set of candidate entities for the target financing project will not be optimized or adjusted, and a letter of intent without prior consent will be signed with each target financing entity in the set of candidate entities for the target financing project. Then, based on the policy label of all projects of the target financing project that are encouraged by the policy, the financing amount range of the target financing project, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence; All projects targeted for financing should apply for policy-based financing that is labeled as "encouraged" under the relevant policy category.

[0086] In this embodiment, the candidate label of the target financing project is "multiple backstops", which means that there are two or more target financing entities in the project candidate entity set that can fully cover the financing needs at the current stage. Even if the approval of a target financing entity fails, there are still other entities to back it up (such as a combination of bank + VC + policy subsidies), and the security of the funding coverage is high. In this embodiment, the existence of a policy-funded project with the policy tag "encouraged" means that at least one policy-funded project (such as the Industrial Internet Special Subsidy or the Innovative Drug R&D Subsidy) has been identified as encouraged based on previous analysis (i.e., the project meets all the characteristics and requirements of the policy and can apply for the policy funding).

[0087] In this embodiment, if the candidate tag for the target financing project is "multiple guarantees" and any one of the policy-funded projects has an "encouraged" policy tag, then the candidate entity set is not optimized. This is because multiple guarantees already ensure funding coverage, and there is "encouraged" policy financing (which can supplement funds or reduce costs). The current candidate set does not need adjustment (e.g., if the original set includes Bank A, VCB, and Financial Leasing C, this set is retained). A letter of intent without preconditions is signed with each target financing entity in the candidate entity set. For example: the bank issues a letter of intent for credit (promising disbursement within 7-14 working days, without government approval); equity investment issues a revocable investment letter of intent (valuation is locked in first, and a decision on whether to formalize the investment is made after due diligence, but not contingent on policy funding); financial leasing / factoring issues a credit line confirmation letter (also not tied to any subsidy trigger). The letter of intent or confirmation letter clearly states that if subsequent policy funds arrive, the high-cost portion can be repaid or replaced in advance, reducing overall funding costs.

[0088] In this embodiment, the candidate label "single catch-all" means that in the set of candidate entities for the project, there is only one financing entity (or a unique combination of entities) that can fully cover the financing needs. If the entity fails to be approved, there are no other alternatives (e.g., only Bank A can lend 15 million, and no other banks or VCs can meet the needs), resulting in low security of fund coverage. In this embodiment, if the candidate label of the target financing project is single-guarantee and there is a policy label of encouragement for any policy financing project, the amount of encouragement policy financing can fill the amount gap of the non-guarantee entity: for example, the project needs 15 million, the minimum investment of sequence 2 (VCC) is 10 million, the gap is 5 million, and the encouragement subsidy can fill 5 million, then VCC is included in the candidate set, and the optimized set is Bank A (single-guarantee) + VCC + subsidy (combined coverage of 15 million), forming a new multi-guarantee structure; In this embodiment, regardless of whether the candidate label has multiple or single coverage options, as long as there is an incentive-based policy financing, applications must be submitted simultaneously, and the application process is parallel to the candidate set adjustment / letter of intent signing. All projects identified in the preliminary analysis as having incentive-based policy financing must be applied for, leaving no available policy resources unrepresented. In this embodiment, after the policy financing is received, the cooperation plan can be adjusted according to the amount. For example, if the original plan was to borrow 15 million from the bank, but the subsidy is 3 million, it can be adjusted to borrow 12 million from the bank to reduce the debt amount and interest costs; or if the original plan was to invest 15 million in VC (10% dilution), but the subsidy is 3 million, it can be adjusted to invest 12 million in VC (8% dilution) to reduce equity dilution.

[0089] The beneficial effects of the above technologies are as follows: by optimizing and adjusting the set of candidate entities for target financing projects based on the candidate tags of target financing projects and the policy tags of all policy-funded projects, financing efficiency can be improved, policy dividends and capital security can be balanced, policy resource utilization can be maximized, and financing costs and interruption risks can be reduced. Example 8:

[0090] This invention provides a financing demand matching system based on in-depth analysis of project materials, used to execute any one of the financing demand matching methods based on in-depth analysis of project materials in Examples 1 to 7, with reference to... Figure 2 ,include: Analysis module: Acquires and analyzes project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage; The determination module acquires financing resource data for the target financing project, and based on the current stage of financing needs and financing resource data of the target financing project, determines the first set of candidate entities and the second set of candidate entities for the target financing project. Candidate Module: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, determine the financing entity sequence, project candidate entity set, and candidate tags of the target financing project; Adjustment Module: Based on the project material data of the target financing project, determine multiple policy financing options for the target financing project and the project policy tags for each policy financing option. Based on the candidate tags of the target financing project and the project policy tags of all policy financing options, optimize and adjust the set of project candidate entities for the target financing project.

[0091] The beneficial effects of the above technologies are as follows: By analyzing the acquired project material data, the current cycle stage of the target financing project and its stage-specific financing needs are determined. Combined with the acquired financing resource data, the first set of candidate entities, the second set of candidate entities, the financing entity sequence, the project candidate entity set, and candidate tags for the target financing project are identified. Based on the project material data, multiple policy-based financing options for the target financing project and the policy tags for each policy-based financing option are determined, and the project candidate entity set for the target financing project is optimized and adjusted. This allows for precise identification of financing needs and matching of financing resources, improving financing efficiency, reducing blind spots and uncertainties in the financing process, balancing policy benefits with capital security, reducing financing costs and risks, increasing the likelihood of successful financing, providing a more comprehensive and scientific financing solution for the target financing project, and promoting the smooth development of the project.

[0092] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for matching financing needs based on in-depth analysis of project materials, characterized in that, include: S1: Obtain and analyze the project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage; S2: Obtain financing resource data for the target financing project, and based on the current stage of financing needs and financing resource data of the target financing project, determine the first candidate entity set and the second candidate entity set for the target financing project; S3: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, determine the financing entity sequence, project candidate entity set, and candidate tags of the target financing project; S4: Based on the project material data of the target financing project, determine multiple policy financing options for the target financing project and the project policy tags for each policy financing option. Optimize and adjust the set of project candidate entities for the target financing project based on the candidate tags of the target financing project and the project policy tags of all policy financing options.

2. The financing demand matching method based on in-depth analysis of project materials according to claim 1, characterized in that, Acquire and analyze project materials data for the target financing project to determine the current stage of the project's lifecycle and its financing needs, including: Obtain project material data for the target financing project, including project market data, project technical data, project financial data, project policy data, and project risk data; Obtain the economic industry classification table, analyze the project market data in the project material data, and combine the economic industry classification table to determine the industry to which the target financing project belongs and the full life cycle of the industry; Based on the project material data, the industry to which the target financing project belongs, and the entire life cycle of the industry, determine the current cycle stage of the target financing project and the primary financing need of the current cycle stage. Based on the project material data of the target financing project, its industry, the entire life cycle of the industry, the current cycle stage, the current financing needs of the current cycle stage, and a series of models, the pre-financing needs of the target financing project for each subsequent cycle stage after the current cycle stage of the entire life cycle of the industry are generated. By taking the pre-financing needs of all subsequent cycle stages after the current cycle stage of the target financing project as constraints based on the full life cycle of the industry to which the target financing project belongs, the first financing need of the current cycle stage of the target financing project is optimized and adjusted to determine the stage financing needs of the current cycle stage of the target financing project.

3. The financing demand matching method based on in-depth analysis of project materials according to claim 1, characterized in that, Obtain financing resource data for the target financing project, including: Obtain financing resource sub-data for multiple target financing entities of the target financing project. The financing resource sub-data includes financing type, financing amount range, matching requirements data, financing cost data, financing conditions data, and historical financing data. Financing types include debt, equity, and other types. Based on the financing resource sub-data of all target financing entities of the target financing project, the financing resource data of the target financing project is determined.

4. The financing demand matching method based on in-depth analysis of project materials according to claim 3, characterized in that, Based on the current stage of financing needs and financing resource data of the target financing project, a first set of candidate entities and a second set of candidate entities are determined, including: Feature extraction is performed on the stage financing needs of the target financing project in the current cycle stage to determine multiple demand features of the target financing project in the current cycle stage and the demand feature value of each demand feature; Feature extraction is performed on the matching requirement data in the financing resource sub-data of each target financing entity of the target financing project to determine multiple matching features for each target financing entity and the matching requirements for each matching feature; Based on each matching feature of each target financing entity, the project material data is extracted to determine the project data for each matching feature of each target financing entity in the target financing project. The matching requirements of each matching feature of each target financing entity are compared with the project data. If the project data of all matching features of the target financing entity meet the matching requirements, the first label of the target financing entity is determined as a candidate entity. If the project data of any matching feature of the target financing entity does not meet the matching requirements, the first label of the target financing entity is determined as a non-candidate entity. Based on all target financing entities whose first label is candidate entity, determine the first candidate entity set for the target financing project; Feature extraction is performed on the financing condition data in the financing resource sub-data of each target financing entity in the first candidate entity set of the target financing project. Multiple condition features of each target financing entity in the first candidate entity set, as well as the condition range and condition label of each condition feature, are determined. The condition label includes performance-based conditions and non-performance-based conditions. Based on the conditional features of each conditional label of each target financing entity in the first candidate entity set that are not performance-based conditions, the project material data is extracted to determine the conditional data of each conditional feature of each target financing entity in the first candidate entity set that are not performance-based conditions. The conditional range and conditional data of each conditional feature of each target financing entity in the first candidate entity set that are not performance-based conditions are compared. If the conditional data of all conditional features of the target financing entity in the first candidate entity set that are not performance-based conditions meet the conditional range, the second label of the target financing entity is determined to be a candidate entity. If the conditional data of any conditional feature of the target financing entity in the first candidate entity set that is not performance-based conditions does not meet the conditional range, the second label of the target financing entity is determined to be a non-candidate entity. Based on the conditional features of each conditional label of each target financing entity in the first candidate entity set as a betting condition, the historical financing data in the financing resource sub-data of each target financing entity in the first candidate entity set is input into the financing prediction model. The predicted conditional value of each conditional label of each target financing entity in the first candidate entity set as a betting condition is determined. It is then determined whether the predicted conditional value of each conditional label of each target financing entity in the first candidate entity set as a betting condition meets the condition range. If the predicted conditional values ​​of all conditional labels of the target financing entity in the first candidate entity set as betting conditions meet the condition range, the third label of the target financing entity is determined as a candidate entity. If the predicted conditional value of any conditional label of the target financing entity in the first candidate entity set as a betting condition does not meet the condition range, the third label of the target financing entity is determined as a non-candidate entity. Based on all target financing entities in the first candidate entity set of the target financing project that are both candidate entities with the second label and the third label, determine the second candidate entity set of the target financing project.

5. The financing demand matching method based on in-depth analysis of project materials according to claim 4, characterized in that, Based on the second candidate entity set of the target financing project, the stage-specific financing needs of the current cycle, and financing resource data, the financing entity sequence, candidate entity set, and candidate tags of the target financing project are determined, including: The financing type and financing cost data of each target financing entity in the second candidate entity set are input into the cost discount model, and the financing cost value of each target financing entity in the second candidate entity set is determined based on the output of the cost discount model. Based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, and the financing cost value and financing type of each target financing entity in the second candidate entity set of the target financing project, calculate the sequence cost value of each target financing entity in the second candidate entity set of the target financing project. Sort the sequence cost values ​​of all target financing entities in the second candidate entity set of the target financing project from smallest to largest to determine the financing entity sequence of the target financing project; Based on all demand characteristics of the current cycle stage of the target financing project, the demand characteristic value of each demand characteristic, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence, calculate the project candidate entity set and candidate labels for the target financing project. Among them, candidate labels include single-backup and multiple-backup.

6. The financing demand matching method based on in-depth analysis of project materials according to claim 1, characterized in that, Based on the project materials data of the target financing project, multiple policy financing options for the target financing project and the project policy tags for each policy financing option are identified, including: The project policy data in the project materials data is broken down to determine the policy-related financing data of multiple policy financing for the target financing project. The policy-related financing data includes the financing amount range, multiple policy characteristics, the policy characteristic requirements of each policy characteristic, and the stage to which the characteristic belongs. Based on all policy financing data, feature extraction was performed on the project market data, project technology data, project financial data, and project risk data in the project material data to determine multiple project characteristics of the target financing project, feature labels for each project characteristic, and project feature values. Among them, feature labels include market characteristics, technology characteristics, financial characteristics, and risk characteristics. Based on all policy features in the policy-related financing data of each policy financing of the target financing project, extract all project features of the target financing project and determine the policy financing vector of each policy financing of the target financing project; Based on the policy feature requirements, feature stage, and project feature value and feature label of each project feature in the policy financing data of each policy financing of the target financing project, the project policy label of each policy financing of the target financing project is determined. The project policy label includes encouragement and restriction.

7. The financing demand matching method based on in-depth analysis of project materials according to claim 6, characterized in that, The set of candidate projects for the target financing project was optimized and adjusted based on the candidate tags of the target financing project and the policy tags of all policy-funded projects, including: If the candidate tag for the target financing project is multiple guarantees and any one of the policy financing projects has an encouragement policy tag, then the set of candidate entities for the target financing project will not be optimized or adjusted, and a letter of intent without prior consent will be signed with each target financing entity in the set of candidate entities for the target financing project. If the candidate label of the target financing project is a single catch-all and there is any policy financing project with the policy label of encouragement, then the set of candidate project entities for the target financing project will be optimized and adjusted based on the range of financing amounts of all projects with the policy label of encouragement for the target financing project, the financing entity sequence of the target financing project, and the financing amount range of each target financing entity in the financing entity sequence. All projects targeted for financing should apply for policy-based financing that is labeled as "encouraged" under the relevant policy category.

8. A financing demand matching system based on in-depth analysis of project materials, characterized in that, A financing demand matching method based on in-depth analysis of project materials, according to any one of claims 1 to 7, includes: Analysis module: Acquires and analyzes project material data of the target financing project to determine the current cycle stage of the target financing project and the stage financing needs of the current cycle stage; The determination module acquires financing resource data for the target financing project, and based on the current stage of financing needs and financing resource data of the target financing project, determines the first set of candidate entities and the second set of candidate entities for the target financing project. Candidate Module: Based on the second candidate entity set of the target financing project, the stage financing needs of the current cycle stage, and financing resource data, determine the financing entity sequence, project candidate entity set, and candidate tags of the target financing project; Adjustment Module: Based on the project material data of the target financing project, determine multiple policy financing options for the target financing project and the project policy tags for each policy financing option. Based on the candidate tags of the target financing project and the project policy tags of all policy financing options, optimize and adjust the set of project candidate entities for the target financing project.

Images