Intelligent recommendation method for financial science and technology products of small and micro enterprises

Abstract

The invention discloses an intelligent recommendation method for financial science and technology products of small and micro enterprises, belongs to the field of financial service recommendation methods, and aims to deeply analyze and construct maps in the financial field, enable recommendation to be more accurate, break through the traditional limitation, adapt to various dynamic requirements of the enterprises, integrate the maps through system acquisition and advanced technologies on cross-domain knowledge fusion and improve the recommendation efficiency. Application accuracy is enhanced, a multi-element recommendation perspective is provided, a quantitative evaluation model and an optimization algorithm are used for optimizing product combinations, key indexes and individual requirements are considered, parameters are dynamically adjusted and strictly verified, the combinations are ensured to be effective and reliable in various market environments, and finally visual recommendation, detailed description and a feedback mechanism are used for improving user experience. The method is improved continuously, the financial service quality and efficiency of small and micro enterprises are improved powerfully, and the healthy development of the enterprises is promoted.

Description

Technical Field

[0001] The present invention relates to the field of financial service recommendation methods, and more specifically, to an intelligent recommendation method for financial technology products of small and micro enterprises. Background Art

[0002] As the status of small and micro enterprises in the national economy becomes increasingly important, the satisfaction of their financial needs is crucial for the survival and development of the enterprises.

[0003] In the field of financial technology product recommendation, the existing recommendation methods mainly recommend financial technology products from two aspects. On the one hand, starting from the financial field, relying on the historical transaction data, financial statements, etc. of small and micro enterprises for analysis and recommendation, but in the process of analysis and recommendation, it is less adaptable to fully consider the complex interrelationships between financial products and the diverse and dynamically changing financial needs of small and micro enterprises, and it is not very convenient to use. On the other hand, although cross-domain knowledge is applied to the financial product recommendation method, it lacks systematicness and depth. For example, there is a research on a recommendation method that simply analogizes the concept of biological evolution, but it does not deeply explore the principles behind the complex relationships of species competition and cooperation in the ecosystem, resulting in the integration of financial field and ecosystem field knowledge in the construction of the cross-domain knowledge graph not being comprehensive and accurate enough, leading to the knowledge graph being unable to accurately reflect the internal connection between the two fields, and thus affecting the accuracy and effectiveness of the recommendation.

[0004] In summary, the existing recommendation methods for financial technology products of small and micro enterprises have certain deficiencies and are difficult to meet the growing complex financial needs of small and micro enterprises. There is an urgent need for a more scientific, comprehensive, and accurate intelligent recommendation method to improve the quality and efficiency of financial services and promote the healthy development of small and micro enterprises. Therefore, we propose an intelligent recommendation method for financial technology products of small and micro enterprises to solve the above existing problems. Summary of the Invention

[0005] Aiming at the problems existing in the prior art, the purpose of the present invention is to provide an intelligent recommendation method for financial technology products of small and micro enterprises, which deeply analyzes and constructs a knowledge graph in the financial field to make the recommendation more accurate, breaks through traditional limitations, and adapts to the diverse and dynamic needs of enterprises. In cross-domain knowledge integration, it integrates the knowledge graph through systematic collection and advanced technology, enhances the application accuracy, provides multiple recommendation perspectives, and quantifies the evaluation model and optimizes the algorithm to optimize the product portfolio, considering key indicators and individual needs, dynamically adjusts parameters and strictly verifies to ensure that the combination is effective and reliable in various market environments. Finally, intuitive recommendation, detailed description, and feedback mechanism improve the user experience, contribute to the continuous improvement of the method, and effectively promote the improvement of the quality and efficiency of financial services for small and micro enterprises and their healthy development.

[0006] To solve the above problems, the present invention adopts the following technical solutions.

[0007] An intelligent recommendation method for fintech products of small and micro enterprises, including the following interrelated and sequential steps.

[0008] S1. In-depth analysis of financial domain knowledge and construction of knowledge graph: For the financial domain, systematically collect various financial product information, where the financial product information covers credit products, insurance products, payment and settlement products, and investment and financial management products, and for each piece of the financial product information, carefully sort out its multi-dimensional features of interest rate pricing mechanism, repayment method, term structure, applicable small and micro enterprise industry types, enterprise scale limits, risk rating systems, and regulatory compliance requirements, and use graph database technology to construct a financial domain knowledge sub-graph. The nodes of the financial domain knowledge sub-graph represent various financial products and their related attributes, and the edges of the financial domain knowledge sub-graph represent the logical relationships between products and attributes and between different products.

[0009] S2. Collection and structured representation of ecological domain knowledge: Systematically collect knowledge about species competition and cooperation relationships in the ecosystem field, including the distribution of different species in ecological niches, resource acquisition strategies, energy flow and material cycle laws in the ecosystem, and population dynamic change mechanisms. The resource acquisition strategies include predation, symbiosis, and parasitism. Use the Web Ontology Language technology (OWL) to structurally represent the knowledge of species competition and cooperation relationships, and construct an ecological domain knowledge sub-graph. The nodes of the ecological domain knowledge sub-graph correspond to different species, ecological processes, and related ecological factors, and the edges of the ecological domain knowledge sub-graph represent the interaction relationships between species and the associations between ecological processes and various elements.

[0010] S3. Fusion and association of cross-domain knowledge graphs: With the help of semantic analysis means in natural language processing technology, including lexical semantic analysis, syntactic dependency analysis, and semantic role annotation, deeply understand the semantic connotations of financial domain and ecological domain knowledge. On this basis, then adopt a knowledge association algorithm based on machine learning to establish a mapping relationship between the financial domain knowledge sub-graph and the ecological domain knowledge sub-graph, and integrate to form a cross-domain knowledge graph. The cross-domain knowledge graph not only intuitively reflects the internal connection of knowledge in the two domains, but also provides a solid data foundation for subsequent knowledge migration and fusion.

[0011] S4. Analogical mapping of financial needs and ecological resources: Make an analogy between the financial needs faced by small and micro enterprises and the resource needs of species in the ecosystem. Regard various pieces of the financial product information as different species in the ecosystem. Each piece of the financial product information meets the specific financial needs of small and micro enterprises through its unique multi-dimensional features, that is, species obtain the required resources through their own survival strategies. The financial needs include capital financing needs, risk aversion needs, and capital management needs, and the resource needs include food, habitat, and sunlight.

[0012] S5. Construction of a Quantitative Evaluation Model for Competitive and Complementary Relationships: Based on the theories and mechanisms of species competition and cooperation in an ecosystem, a quantitative evaluation model for the competitive and complementary relationships of the financial product information is constructed. The quantitative evaluation model for the competitive and complementary relationships also fully considers multiple key attribute indicators of the financial product information to improve the accuracy of the construction of the quantitative evaluation model for the competitive and complementary relationships. The key attribute indicators include profitability indicators, risk indicators, liquidity indicators, and cost indicators. The quantitative evaluation model for the competitive and complementary relationships of the financial product information also combines the personalized demand characteristics of small and micro enterprises, and through the multi-criteria decision-making analysis method of the Analytic Hierarchy Process (AHP), assigns corresponding weights to different demand dimensions of small and micro enterprises, and quantitatively evaluates the competitive and complementary relationships of different financial product information when meeting the various financial needs of small and micro enterprises.

[0013] S6. Search for Product Combinations Based on Optimization Algorithms: Using intelligent optimization algorithms with the maximization of the comprehensive financial demand satisfaction of small and micro enterprises, the minimization of financial costs, risk diversification, and revenue stability as multi-objective functions, search in the cross-domain knowledge graph to generate potential financial product information combinations based on the cross-domain knowledge graph. The intelligent optimization algorithms include Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing Algorithm (SA). And during the search process, the intelligent optimization algorithms dynamically adjust the composition parameters of the financial product information combination. The composition parameters include product types, product quantities, and the weight distribution of each product in the combination to adapt to the dynamically changing financial needs of different small and micro enterprises at different development stages and market environments.

[0014] S7. Rationality Verification of Product Combinations: Conduct a comprehensive rationality verification on the generated financial product information combination. The rationality verification includes carefully examining whether there are conflicts or incompatibilities in the application conditions, usage periods, operation processes, and risk characteristics of the financial product information combination from the perspective of product compatibility. The rationality verification also includes carrying out a market adaptability assessment, and combining current macroeconomic situations, financial market fluctuations, industry development trends, and regulatory policy changes to analyze the feasibility and potential benefits of the financial product information combination in the real market environment. The rationality verification also includes backtesting analysis of historical data and prediction simulation of future market trends to evaluate the performance of the financial product information combination under different market scenarios, ensuring that the financial products recommended to small and micro enterprises not only meet the actual needs of the enterprises but also can operate steadily in the market.

[0015] S8. Output and feedback of recommendation results: The financial product information combination verified in the above step S7 is recommended to small and micro enterprise users in an intuitive and easy-to-understand manner, and a detailed financial product description is provided. Then a user feedback mechanism is established. The financial product description includes an introduction to the functions of the financial products, the advantages of the combination, and potential risk warning information. The user feedback mechanism collects the experience and opinions of small and micro enterprises in the process of using the recommended financial product information combination, so as to facilitate the continuous optimization and improvement of the recommendation method in the later stage, so as to continuously improve the accuracy and practicality of the recommendation.

[0016] Furthermore, in step S1, the credit products include credit loans, mortgage loans and supply chain finance loans, the insurance products include property insurance, credit guarantee insurance and loan guarantee insurance, the payment settlement products include electronic payment platforms and account management systems, the investment and financial products include money funds and bond funds, the repayment methods include equal installments of principal and interest, interest first and principal later, and one-time repayment of principal and interest, and the term structure includes short-term, medium-term and long-term.

[0017] Furthermore, in step S3, the lexical semantic analysis uses a pre-trained language model, which includes BERT and GPT, to perform feature extraction on text data in the financial and ecological fields to obtain more accurate word vector representations, thereby capturing the semantic similarity and semantic association between words. The syntactic dependency analysis uses a deep learning model, which is a dependency syntactic analyzer based on the Transformer architecture. It performs syntactic structure analysis on complex domain sentences, clarifies the dependency relationship between each component in the sentence, and provides a basis for the subsequent semantic role labeling. The semantic role labeling uses an end-to-end model based on a neural network. The end-to-end model based on a neural network is an LSTM-CRF model combined with an attention mechanism. On the basis of the syntactic analysis, the semantic role of each predicate in the sentence is accurately identified. The semantic roles include agent, patient, time, and place, and the semantic information of domain knowledge is further understood in depth.

[0018] Further, in the step S3, the knowledge association algorithm based on machine learning is calculated based on a deep learning-based graph neural network model. The deep learning-based graph neural network model is Graph Neural Network. Feature extraction and representation learning are performed on the financial domain knowledge sub-graph and the ecological domain knowledge sub-graph, and a graph matching algorithm is used to calculate the similarity between the two graphs, thereby discovering potential knowledge associations. Then, based on an association mining algorithm that combines rules and statistics, domain-specific rules are first formulated. The domain-specific rules include similarity rules based on the functions of financial products and the survival strategies of ecological species. Then, through statistical analysis methods, the statistical analysis methods include co-occurrence analysis and association rule mining, and knowledge association relationships that conform to the rules are mined from large-scale domain data to achieve the internal effective connection between the two domain knowledge sub-graphs.

[0019] Further, in the step S5, the internal rate of return of the profitability index is accurately calculated through a cash flow discount model, considering the cash inflows and outflows of the financial product during its entire life cycle. The net present value of the profitability index is then calculated by discounting based on the market interest rate and the expected return of the product to reflect the current value of the product and improve the accuracy of the quantitative evaluation model for the competition and complementarity relationship.

[0020] The credit risk of the risk index is evaluated using a big data-based credit scoring model, comprehensively considering the financial status, credit records, and business stability factors of small and micro enterprises. The market risk of the risk index is quantified through a Value at Risk (VaR) model or a Conditional Value at Risk (CVaR) model to measure the maximum loss that a financial product may face at a certain confidence level.

[0021] The liquidity risk of the liquidity index is evaluated by calculating the time required for capital liquidation and the capital turnover efficiency index. The capital liquidation speed of the liquidity index is quantified by analyzing factors such as the trading activity and market depth of the financial product. The capital turnover cycle of the liquidity index is calculated based on the business model of small and micro enterprises and the repayment arrangement of the financial product.

[0022] The interest rate cost in the cost index is accurately calculated according to the interest rate pricing formula of the product, and the handling fee rate in the cost index is obtained by summarizing and statistics of various handling fee items.

[0023] Further, in the step S5, the personalized demand characteristics of small and micro enterprises are realized in the form of a detailed demand survey of small and micro enterprises. The demand survey includes, but is not limited to, questionnaire surveys, face-to-face interviews, and data analysis methods, to collect the degree of attention of small and micro enterprises to various attributes of financial products at different development stages, different industry characteristics, and different business objectives.

[0024] The corresponding weights are assigned to different demand dimensions of small and micro enterprises. Specifically, the analytic hierarchy process (AHP) is used to construct a hierarchical structure model, and the financial demand objectives of small and micro enterprises are decomposed into criteria and indicators at multiple levels. The relative importance between elements at each level is determined through pairwise comparisons, so as to calculate the weight vectors of different demand dimensions. Combining with the objective weighting method of the entropy weight method, the weights are adjusted and optimized to ensure that the weights can not only reflect the subjective demand preferences of small and micro enterprises but also reflect the objective information of the data.

[0025] Furthermore, in step S6, the genetic algorithm (GA) dynamically adjusts the crossover probability and mutation probability parameters. According to the change of the fitness function value during the search process, an adaptive parameter adjustment strategy is adopted. For example, when the fitness of the population individuals tends to be consistent, the mutation probability is appropriately increased to increase the diversity of the population and avoid the algorithm falling into a local optimal solution. The particle swarm optimization algorithm (PSO) dynamically adjusts parameters such as the inertia weight and learning factor according to the change of the flight speed and position of the particles, so that the particles achieve a balance between global search and local search, improving the search efficiency and accuracy of the algorithm. The simulated annealing algorithm (SA) adaptively adjusts the cooling rate according to the change of the current temperature and the objective function value to ensure that the algorithm converges to the global optimal solution or an approximate global optimal solution within a reasonable time.

[0026] Furthermore, in step S7, the macroeconomic situation analysis selects key macroeconomic indicators such as the growth rate of gross domestic product, the inflation rate, and the interest rate level, establishes an econometric model between the macroeconomic indicators and the demand and return of financial products, and evaluates the market adaptability of the financial product information combination by analyzing the impact of changes in macroeconomic variables on the financial product market.

[0027] The financial market volatility is analyzed by using a time series analysis model for the main indicators of the financial market combined with machine learning algorithms to comprehensively judge the future trend of the financial market and provide a basis for the adjustment of the financial product information combination. The time series analysis model uses the ARIMA model. The main indicators of the financial market include stock indexes, bond yields, and exchange rates. The machine learning algorithm is the support vector machine (SVM) algorithm.

[0028] The industry development trend analysis analyzes the market concentration, competition pattern, and threat factors of new entrants in the industry where small and micro enterprises are located, and uses the Porter five force model tool to evaluate the impact of the industry competition environment on the demand for financial products, ensuring that the recommended financial product information combination can adapt to the dynamic changes of industry development.

[0029] Compared with the prior art, the advantages of the present invention are as follows.

[0030] (1) This solution accurately meets financial needs. In terms of in-depth analysis of financial domain knowledge and construction of knowledge graphs, it comprehensively sorts out multi-dimensional features of financial products to construct knowledge sub-graphs, changing the limitation of relying solely on historical transaction data and financial statement analysis for recommendations in the past. It can better adapt to the diverse and dynamically changing financial needs of small and micro enterprises, effectively improving the accuracy of recommendations. For example, by carefully analyzing attributes such as interest rate pricing and repayment methods of various products such as credit, insurance, payment and settlement, and investment financial products, products can be accurately matched to the enterprise's capital financing, risk aversion, and capital management needs at specific stages.

[0031] (2) This solution enhances the accuracy of cross-domain knowledge application. By systematically collecting ecological domain knowledge and representing it in a structured manner, and then using advanced natural language processing and machine learning technologies to fuse and form a cross-domain knowledge graph, it overcomes the problems of lack of systematicness and depth in cross-domain knowledge application in existing methods, accurately integrating financial and ecological domain knowledge, making the recommendation basis more scientific and comprehensive. For example, based on an in-depth understanding of the relationships between ecological species and financial products, potential product associations can be mined, providing a richer perspective for recommendations.

[0032] (3) This solution optimizes product portfolio and evaluation. The constructed quantitative evaluation model for the competitive and complementary relationships of financial products fully considers key indicators such as profitability, riskiness, liquidity, and costliness, and assigns weights in combination with the personalized demand characteristics of small and micro enterprises, solving the defect that previous recommendations did not deeply analyze the complex relationships of products. Using intelligent optimization algorithms to search for product portfolios, the combination parameters can be dynamically adjusted according to enterprise needs, achieving multi-objective optimization such as maximizing the overall financial demand satisfaction. And through various aspects of rationality verification, it ensures that the product portfolio not only conforms to the actual situation of the enterprise but also can operate stably in different market environments, greatly improving the effectiveness and reliability of recommendations.

[0033] (4) This solution improves user experience and continuously improves the method. It directly recommends the verified product portfolio to small and micro enterprises and provides detailed explanations. At the same time, a user feedback mechanism is established, which is convenient for enterprises to understand products and helps to continuously optimize the recommendation method, continuously improving the accuracy and practicality of recommendations, better serving the financial needs of small and micro enterprises, and promoting their healthy development. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the method step architecture of the present invention.

[0035] Figure 2 It is a mind map of step S1 of the present invention.

[0036] Figure 3 It is a mind map of step S2 of the present invention.

[0037] Figure 4It is the mind map of step S3 of the present invention.

[0038] Figure 5 It is the mind map of step S4 of the present invention.

[0039] Figure 6 It is the mind map of step S5 of the present invention.

[0040] Figure 7 It is the mind map of step S6 of the present invention.

[0041] Figure 8 It is the mind map of step S7 of the present invention.

[0042] Figure 9 It is the mind map of step S8 of the present invention. Specific embodiments

[0043] Next, in combination with the accompanying drawings of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and completely; obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] Embodiment 1

[0045] In combination with the accompanying drawings of the specification Figures 1-9 , the process of recommending fintech products for a small and micro enterprise in the manufacturing industry.

[0046] S1. In-depth analysis and map construction of financial domain knowledge.

[0047] For this small and micro enterprise, first, information collection in the financial field is carried out. In terms of credit products, it includes credit loans provided by banks, whose interest rate pricing fluctuates within a certain range according to the credit rating of the small and micro enterprise, the repayment method is equal principal and interest, and the term structure is medium-term (1 - 3 years); mortgage loans use the enterprise's fixed assets as collateral, with relatively low interest rates, the repayment method is interest first and principal later, and the term is long-term (3 - 5 years); insurance products include property insurance, which protects the losses of the enterprise's fixed assets in case of accidents such as natural disasters; credit guarantee insurance is used to increase the credit limit when the enterprise applies for a loan; for payment and settlement products, a commonly used electronic payment platform is selected to facilitate the enterprise's daily transaction settlement; the investment and financial management product is a money fund, which has the characteristics of strong liquidity and low risk.

[0048] Using graph database technology, these financial products and their related attributes are constructed into a knowledge sub-graph in the financial domain. Specifically, the credit loan node is connected to attribute nodes such as its interest rate range, equal principal and interest repayment method, medium term, and the scale limit of applicable small and micro manufacturing enterprises. The edges represent the logical relationships between them, such as the interest rate of the credit loan being affected by the credit rating of small and micro enterprises.

[0049] S2. Collection and structured representation of ecological domain knowledge.

[0050] Collect knowledge in the ecosystem domain. In the ecological environment where this small and micro manufacturing enterprise is located, there are competition and cooperation relationships among different species. For example, in the ecological area around its factory, there is an insect population that feeds on the factory waste (similar to the predation in the resource acquisition strategy). These insects have a symbiotic relationship with the surrounding plants. The insects help the plants pollinate, and the plants provide habitats for the insects. At the same time, there is a carnivore population that preys on small mammals, and there is a competition relationship between them for the same food resources.

[0051] Using the Web Ontology Language technology (OWL) to structurally represent this knowledge of species competition and cooperation relationships, and construct an ecological domain knowledge sub-graph. For example, the insect population node is connected to the plant population node by a symbiotic relationship edge and to the carnivore population node by a competition relationship edge, and is also connected to ecological processes such as energy flow (insects obtain energy from plants, and carnivores obtain energy from insects) and related ecological factor nodes (such as habitat type, etc.).

[0052] S3. Fusion and association of cross-domain knowledge graphs.

[0053] With the help of natural language processing technology, use the pre-trained BERT model to perform lexical semantic analysis on the text data in the financial and ecological domains, and extract the word vector representations of texts related to financial products and ecological species. For example, the "funds financing" function of the credit loan has a certain semantic similarity to the "energy acquisition" of insects to obtain food resources. This association can be initially discovered through the calculation of the similarity of word vectors.

[0054] Use a dependency parser based on the Transformer architecture to perform syntactic dependency analysis on complex sentences, such as "Small and micro enterprises apply for credit loans for production capital turnover, just like insects looking for food to survive", to clarify the dependency relationships of each component in the sentence and provide a basis for semantic role annotation.

[0055] The LSTM-CRF model is combined with the attention mechanism for semantic role labeling to further understand the semantic information of domain knowledge. On this basis, the Graph Neural Network (GNN) based on deep learning is used to extract features and perform representation learning on the financial domain knowledge sub-graph and the ecological domain knowledge sub-graph. The similarity is calculated through the graph matching algorithm to discover potential knowledge associations. At the same time, rules based on the similarity between the functions of financial products and the survival strategies of ecological species are formulated. For example, risk-averse financial products are associated with ecological species with defense mechanisms. Then, statistical methods such as co-occurrence analysis are used to mine knowledge association relationships that conform to the rules, and finally an integrated cross-domain knowledge graph is formed.

[0056] S4. Analogical mapping between financial needs and ecological resources.

[0057] This small and micro manufacturing enterprise faces the need for financing to expand production scale and purchase new equipment, which is similar to the large demand of insect populations for food resources in the ecosystem to maintain population growth. The enterprise also has a need for risk aversion, worrying that market fluctuations will affect product sales and lead to a broken capital chain, similar to the need of some species in the ecosystem for a stable habitat to avoid natural enemies and harsh environments. The need for fund management is like the need of species in the ecosystem for reasonable allocation and utilization of energy. Regarding various financial products as different species in the ecosystem, credit loans are like resources that can provide a large amount of food and can meet the enterprise's financing needs. Property insurance is like a safe habitat, providing risk aversion protection for the enterprise. Money funds are similar to an efficient energy management mechanism, helping the enterprise to reasonably manage idle funds.

[0058] S5. Construction of a quantitative evaluation model for competition and complementary relationships.

[0059] For the profitability index, the internal rate of return of the credit loan is calculated through the cash flow discount model, considering the enterprise's repayment cash flow and interest expenses during the loan term. The net present value is obtained by discounting based on the current market interest rate and expected returns. For example, after calculation, the internal rate of return of the credit loan applied by this enterprise is 8%, and the net present value is positive, indicating that it has a certain attraction in terms of profitability.

[0060] Regarding the risk index, credit risk is evaluated using a big data-based credit scoring model, integrating information such as the enterprise's financial statements and tax records. The credit score of this enterprise is good, and the credit risk is at a relatively low level. Market risk is quantified through the VaR model at a 95% confidence level. After analysis, under market fluctuations, the probability of the maximum loss faced by the enterprise's loan is relatively small.

[0061] Among the liquidity indicators, the time required for the capital of the credit loan to be liquidated is gradually liquidated according to the repayment plan within the loan term. The capital turnover efficiency is calculated based on the production cycle and repayment cycle of the enterprise. The capital turnover cycle is one year, and the capital liquidation speed is at a medium level by analyzing the trading activity and market depth of similar loans in the financial market.

[0062] Among the cost indicators, the interest rate cost of the credit loan is accurately calculated as 6% per annum according to the bank's interest rate pricing formula, and the handling fee rate is 1% of the loan amount.

[0063] By means of questionnaire surveys and data analysis, the personalized demand characteristics of this small and micro enterprise are understood. It is found that the enterprise pays more attention to the timeliness of capital financing and cost control at the current development stage. The analytic hierarchy process (AHP) is used to construct a hierarchical structure model, and the financial demand goal is decomposed into criteria such as capital financing, risk aversion, and capital management. The relative importance of each hierarchical element is determined through pairwise comparisons. The weight of the capital financing demand dimension is calculated as 0.5, the weight of the risk aversion demand dimension is 0.3, and the weight of the capital management demand dimension is 0.2. Combined with the entropy weight method for adjustment and optimization, the weight vector is finally determined, so as to quantitatively evaluate the competition and complementary relationship of different financial products when meeting the enterprise's various financial needs.

[0064] S6. Product portfolio search based on optimization algorithms.

[0065] The genetic algorithm (GA) is used to search in the cross-domain knowledge graph with the maximization of the comprehensive financial demand satisfaction of small and micro enterprises, the minimization of financial costs, the diversification of risks, and the stability of returns as multi-objective functions. During the search process, the crossover probability and mutation probability parameters are dynamically adjusted according to the change of the fitness function value. For example, the initial crossover probability is set to 0.8, and the mutation probability is 0.1. When the fitness of the population individuals tends to be consistent, the mutation probability is increased to 0.3 to increase the population diversity and avoid falling into local optimal solutions. After multiple iterations, potential financial product information combinations are generated, such as a combination of a 500,000 yuan credit loan, a 2 million yuan property insurance coverage, and a 300,000 yuan money fund investment, and the weight distribution of each product in the combination is determined. The weight of the credit loan is 0.5, the weight of the property insurance is 0.3, and the weight of the money fund is 0.2.

[0066] S7. Rationality verification of the product portfolio.

[0067] From the perspective of product compatibility review, the enterprise has met the application conditions of the credit loan. The usage period matches the enterprise's production equipment procurement and repayment plan, the enterprise can master the operation process proficiently, and the risk characteristics are complementary to those of property insurance and money funds, without conflicts or incompatibilities.

[0068] Conduct a market adaptability assessment, analyze the current macroeconomic situation. The growth rate of the gross domestic product is stable at 6%, the inflation rate is controlled at 3%, and the interest rate level is moderate. Through the establishment of an econometric model analysis, it is concluded that the financial product information portfolio has good adaptability in the current macroeconomic environment, with stable expected returns. Use the ARIMA model combined with the support vector machine (SVM) algorithm to analyze the fluctuations in the financial market. The fluctuations in the stock index and bond yields are within a reasonable range, and the exchange rate is stable, with little impact on this portfolio. Analyze the development trend of the manufacturing industry through the Porter five force model. The market concentration is moderate, the competition pattern is stable, and the threat of new entrants is small. This portfolio can meet the financial needs of enterprises in the industry development. At the same time, conduct a backtest analysis on historical data to simulate the performance of enterprises using this portfolio in past similar market scenarios. The results show that the financial condition and business stability of enterprises have been effectively improved. The prediction simulation of future market trends also indicates that this portfolio is feasible in the next period of time.

[0069] S8. Output and feedback of the recommendation results.

[0070] Recommend the verified financial product information portfolio to this small and micro enterprise user, and provide detailed descriptions of financial products. For example, explain to the enterprise the amount, interest rate, repayment method of the credit loan and its advantage of quickly solving the problem of capital financing, the scope of protection, claim settlement process of property insurance and how to reduce enterprise risks, the income characteristics, liquidity advantages of money funds and their management functions for idle funds. At the same time, establish a user feedback mechanism to collect their experiences and opinions during the process of enterprises using the recommended product portfolio. For example, whether the enterprise feedback that the credit loan approval process can be further simplified, whether the income of money funds can be more stable, etc., so as to continuously optimize and improve the recommendation method in the later stage.

[0071] Embodiment 2

[0072] Combined with the accompanying drawings of the specification Figures 1-9 , the recommendation process of financial technology products for a small and micro enterprise of science and technology service type.

[0073] S1. In-depth analysis and map construction of financial domain knowledge.

[0074] For small and micro enterprises of science and technology service type, the credit products include special loans for high-tech enterprises with preferential interest rates, the repayment method is lump-sum repayment of principal and interest, and the term structure is short-term (6 months - 1 year) to meet the short-term capital needs of enterprise R & D projects. The insurance products include science and technology insurance to protect the enterprise from technical risks and intellectual property risks during the R & D process. The payment and settlement products select an account management system with cross-border payment functions to facilitate the business transactions between enterprises and international customers. The investment and financial management products are bond funds with relatively stable returns, which are suitable for enterprises to preserve and increase the value of funds.

[0075] When constructing the knowledge sub - graph in the financial field, the special loan node for high - tech enterprises is connected to attribute nodes such as preferential interest rates, the method of repaying the principal and interest in one lump sum, short - term periods, and the scale and industry types of applicable small and micro technology - service enterprises, clarifying the logical relationships between various products and attributes.

[0076] S2. Collection and structured representation of ecological domain knowledge.

[0077] In the ecological environment where small and micro technology - service enterprises are located, there is a competitive relationship between innovative enterprises and traditional enterprises, similar to different species competing for ecological niches in an ecosystem. For example, within the same science and technology park, this small and micro enterprise competes with other enterprises of the same type for limited human resources and market share. At the same time, there are cooperative relationships with upstream and downstream enterprises. For example, it cooperates with scientific research institutions to obtain technical support (similar to a symbiotic relationship) and cooperates with suppliers to ensure raw material supply (similar to a predatory relationship in the resource acquisition strategy).

[0078] Use OWL technology to construct the ecological domain knowledge sub - graph, taking small and micro technology - service enterprises and their competitive and cooperative enterprises as nodes, representing relationships such as competition and cooperation through different types of edges, and connecting relevant ecological process and ecological factor nodes, such as ecological process nodes like talent flow and technological innovation.

[0079] S3. Fusion and association of cross - domain knowledge graphs.

[0080] Adopt the BERT model for lexical semantic analysis to mine the semantic associations between texts in the science and technology finance field and the ecological field. For example, the "risk protection" function of science and technology insurance has semantic similarity with the "self - protection" mechanism of species in the ecosystem. Use a deep - learning dependency parser to analyze complex sentences, such as "Small and micro technology - service enterprises purchase science and technology insurance to resist R & D risks, just as species develop defense mechanisms in biological evolution to cope with environmental changes", determine the sentence structure relationship, perform semantic role annotation through an end - to - end neural network model to further understand semantic information, and based on deep - learning graph matching algorithms and association mining algorithms that combine rules and statistics, integrate the financial domain knowledge sub - graph and the ecological domain knowledge sub - graph, establish mapping relationships, and form a cross - domain knowledge graph. For example, it is found that there is an association between science and technology insurance and ecological species with defense strategies, and this knowledge association is strengthened by statistically analyzing the co - occurrence of technology enterprises and relevant ecological factors.

[0081] S4. Analogical mapping between financial needs and ecological resources.

[0082] The financing needs of this technology service-oriented small and micro enterprise are mainly used for R & D investment and market expansion, similar to the needs of species in an ecosystem for key resources in order to evolve and expand their populations. The risk aversion needs stem from the uncertainty of technology R & D and the fierceness of market competition, just like the needs of species in an ecosystem for a safe environment when facing environmental changes and threats from natural enemies. The fund management needs are to improve the efficiency of fund use, similar to the efficient use of energy by species in an ecosystem.

[0083] By analogy, financial products are regarded as ecological resources. Special loans for high-tech enterprises are the key resources to meet R & D financing, science and technology insurance is a safety guarantee to cope with risks, and bond funds are effective tools to optimize fund management.

[0084] S5. Construction of a quantitative evaluation model for competition and complementary relationships.

[0085] In terms of profitability indicators, the internal rate of return of bond funds is calculated to be 7% through the discounted cash flow model according to the income situation of its investment portfolio and market interest rate fluctuations. The net present value is calculated by discounting the expected income and market interest rate, reflecting its value in the current market environment.

[0086] Among the risk indicators, the risk assessment of science and technology insurance comprehensively considers factors such as the technology R & D risk and intellectual property risk of the enterprise, and is quantified through a professional risk assessment model. The credit risk is based on the enterprise's credit record and financial condition, and is evaluated using a big data credit scoring model. The enterprise has good credit and the risk is within a controllable range. The market risk is analyzed through the CVaR model at a 90% confidence level to determine the potential loss risk faced by the enterprise during market fluctuations.

[0087] Among the liquidity indicators, the cash conversion speed of bond funds is analyzed according to its trading activity and market depth, and has a certain liquidity. The fund turnover cycle is calculated according to the enterprise's fund use plan and the redemption regulations of the fund, and is 1.5 years.

[0088] Among the cost indicators, the interest rate cost of special loans for high-tech enterprises is determined to be 5% per annum according to government support policies and bank pricing formulas, and the handling fee rate is relatively low. By conducting face-to-face interviews and data analysis to understand the personalized demand characteristics of the enterprise, the enterprise pays more attention to risk aversion and fund management at the current stage. Using the analytic hierarchy process (AHP) to construct a hierarchical structure model, the weight of the risk aversion demand dimension is determined to be 0.4, the weight of the fund management demand dimension is 0.4, and the weight of the financing demand dimension is 0.2, and it is adjusted and optimized in combination with the entropy weight method to quantitatively evaluate the competition and complementary relationships of financial products.

[0089] S6. Product portfolio search based on optimization algorithms.

[0090] Using the Particle Swarm Optimization (PSO) algorithm, search in the cross-domain knowledge graph with the maximization of the enterprise's comprehensive financial demand satisfaction as multiple objective functions. Dynamically adjust parameters such as the inertia weight and learning factors according to the changes in the flight speed and position of the particles. For example, the initial inertia weight is 0.9, and the learning factors are 1.5 and 1.6 respectively. As the search process progresses, when it is found that the particles are overly concentrated in a local area, appropriately reduce the inertia weight to 0.7 and increase the learning factors to 2 and 2, so that the particles achieve a balance between global search and local search. After algorithm iteration, generate potential financial product combinations, such as a special loan of 800,000 yuan for high-tech enterprises, a science and technology insurance coverage of 3 million yuan, and a bond fund investment of 500,000 yuan, and determine the weight distribution of each product in the combination. The weight of the special loan is 0.4, the weight of the science and technology insurance is 0.4, and the weight of the bond fund is 0.2.

[0091] S7. Rationality verification of the product combination.

[0092] In terms of product compatibility, the enterprise meets the application conditions for the special loan, the usage period matches the R & D project cycle, the operation process is convenient, and the risk diversification effect is good after combining with science and technology insurance and bond funds. In the market adaptability assessment, analyze the macroeconomic situation. The domestic GDP growth rate is relatively high, the science and technology innovation industry is supported by policies, and the interest rate level is conducive to enterprise financing. This combination has advantages in the macroeconomic environment. Analyze the financial market volatility through the time series analysis model and machine learning algorithm. The science and technology financial market is stable and has little impact on the combination. Use the Porter Five Force Model to analyze the development trend of the science and technology service industry. The industry is growing rapidly, the competition is fierce but there are many innovation opportunities. This combination can meet the financial needs of enterprises in the industry development. The backtesting of historical data and the simulation of future market trend prediction show that enterprises adopting this combination have maintained a good financial condition in the past market fluctuations and are also expected to achieve stable development in the future.

[0093] S8. Output and feedback of the recommendation results.

[0094] Recommend the verified financial product combination to small and micro technology service enterprises, provide detailed product descriptions, including preferential policies for special loans, coverage and claims settlement processes for science and technology insurance, investment strategies and expected returns for bond funds, etc. Establish a user feedback mechanism to collect the opinions of enterprises during the use process, such as feedback information on whether the claims settlement speed of science and technology insurance can be improved and investment suggestions for bond funds, so as to optimize and improve the recommendation method.

[0095] Through the above embodiments, the specific application process and effects of the intelligent recommendation method for fintech products of small and micro enterprises can be more clearly demonstrated, further verifying the feasibility and effectiveness of the method. In actual applications, it can be flexibly adjusted and optimized according to the specific situations of different small and micro enterprises to meet their diverse financial needs.

[0096] The above is only a preferred specific embodiment of the present invention; however, the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution of the present invention and its improved concept, making equivalent replacements or changes, should be covered by the protection scope of the present invention.

Claims

1. A method for intelligently recommending financial technology products for small and micro enterprises, characterized in that: It includes the following interrelated and sequential steps: S1. In-depth analysis and graph construction of financial knowledge: In the financial field, systematically collect information on various financial products, including credit products, insurance products, payment and settlement products, and investment and financial products. The information on each financial product is sorted out in detail, including its interest rate pricing mechanism, repayment method, term structure, applicable small and micro enterprise industry types, enterprise size restrictions, risk rating system, and multi-dimensional characteristics of regulatory compliance requirements. Use graph database technology to construct a sub-graph of financial knowledge. The nodes of the sub-graph represent various financial products and their related attributes, and the edges of the sub-graph represent the logical relationship between products and attributes and between different products. S2. Collection and structured representation of ecological domain knowledge: systematically collect knowledge about species competition and cooperation in the field of ecosystems, including the distribution of different species in ecological niches, resource acquisition strategies, energy flow and material circulation laws in ecosystems, and population dynamic change mechanisms. The resource acquisition strategies include predation, symbiosis, and parasitism. Use the network ontology language technology (OWL) to structure the knowledge of species competition and cooperation, and construct an ecological domain knowledge sub-graph. The nodes of the ecological domain knowledge sub-graph correspond to different species, ecological processes, and related ecological factors. The edges of the ecological domain knowledge sub-graph represent the interaction between species and the relationship between ecological processes and various elements. S3. Fusion and association of cross-domain knowledge graphs: With the help of semantic analysis methods in natural language processing technology, including lexical semantic analysis, syntactic dependency analysis, and semantic role labeling, we deeply understand the semantic connotation of knowledge in the financial field and the ecological field. On this basis, we use the knowledge association algorithm based on machine learning to establish the mapping relationship between the financial field knowledge sub-graph and the ecological field knowledge sub-graph, and integrate them to form a cross-domain knowledge graph. The cross-domain knowledge graph not only intuitively reflects the inherent connection between the knowledge of the two fields, but also provides a solid data foundation for subsequent knowledge migration and fusion; S4. Analogy mapping of financial needs and ecological resources: The financial needs faced by small and micro enterprises are compared with the resource needs of species in the ecosystem. The various types of financial product information are regarded as different species in the ecosystem. Each type of financial product information meets the specific financial needs of small and micro enterprises through its unique multi-dimensional characteristics, that is, species obtain the required resources through their own survival strategies. The financial needs include financing needs, risk avoidance needs and fund management needs, and the resource needs include food, habitat and sunlight; S5. Construction of a quantitative evaluation model for competition and complementary relationships: Based on the theory and mechanism of competition and cooperation between species in an ecosystem, a quantitative evaluation model for the competition and complementary relationship of the financial product information is constructed. The quantitative evaluation model for the competition and complementary relationship of the financial product information also fully considers multiple key attribute indicators of the financial product information to improve the accuracy of the construction of the quantitative evaluation model for the competition and complementary relationship. The key attribute indicators include profitability indicators, risk indicators, liquidity indicators, and cost indicators. The quantitative evaluation model for the competition and complementary relationship of the financial product information is also combined with the personalized demand characteristics of small and micro enterprises. Through the multi-criteria decision analysis method of the analytic hierarchy process (AHP), corresponding weights are assigned to different demand dimensions of small and micro enterprises, and the competition and complementary relationship of different financial product information in meeting the various financial needs of small and micro enterprises is quantitatively evaluated; S6. Product portfolio search based on optimization algorithm: Using intelligent optimization algorithm to maximize the satisfaction of comprehensive financial needs of small and micro enterprises, minimize financial costs, diversify risks and stabilize returns as multi-objective functions, search in the cross-domain knowledge graph to generate potential financial product information combinations based on the cross-domain knowledge graph, the intelligent optimization algorithm includes genetic algorithm (GA), particle swarm optimization algorithm (PSO) and simulated annealing algorithm (SA), and in the search process, the intelligent optimization algorithm dynamically adjusts the composition parameters of the financial product information combination, the composition parameters include product types, product quantity and weight distribution of each product in the combination, so as to adapt to the dynamically changing financial needs of different small and micro enterprises at different development stages and market environments; S7. Rationality verification of product portfolio: Conduct a comprehensive rationality verification on the generated financial product information portfolio. The rationality verification includes carefully examining whether there are conflicts or incompatibilities in the application conditions, usage period, operation process and risk characteristics of the financial product information portfolio from the perspective of product compatibility. The rationality verification also includes conducting a market adaptability assessment, combining the current macroeconomic situation, financial market fluctuations, industry development trends and regulatory policy changes to analyze the feasibility and potential benefits of the financial product information portfolio in the actual market environment. The rationality verification also includes backtesting analysis of historical data and forecast simulation of future market trends to evaluate the performance of the financial product information portfolio in different market scenarios to ensure that the financial products recommended to small and micro enterprises not only meet the actual needs of the enterprises, but also operate stably in the market; S8. Output and feedback of recommendation results: The financial product information combination verified in the above step S7 is recommended to small and micro enterprise users in an intuitive and easy-to-understand manner, and a detailed financial product description is provided. Then a user feedback mechanism is established. The financial product description includes an introduction to the functions of the financial products, the advantages of the combination, and potential risk warning information. The user feedback mechanism collects the experience and opinions of small and micro enterprises in the process of using the recommended financial product information combination, so as to facilitate the continuous optimization and improvement of the recommendation method in the later stage, so as to continuously improve the accuracy and practicality of the recommendation.

2. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S1, the credit products include credit loans, mortgage loans and supply chain financial loans, the insurance products include property insurance, credit guarantee insurance and loan guarantee insurance, the payment settlement products include electronic payment platforms and account management systems, the investment and financial products include money funds and bond funds, the repayment methods include equal installments of principal and interest, interest first and principal later, and one-time repayment of principal and interest, and the term structure includes short-term, medium-term and long-term.

3. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S3, the lexical semantic analysis adopts a pre-trained language model, which includes BERT and GPT, to perform feature extraction on text data in the financial and ecological fields to obtain more accurate word vector representation, thereby capturing the semantic similarity and semantic association between words. The syntactic dependency analysis uses a deep learning model, which is a dependency syntactic analyzer based on the Transformer architecture. It performs syntactic structure analysis on complex domain sentences, clarifies the dependency relationship between each component in the sentence, and provides a basis for the subsequent semantic role labeling. The semantic role labeling uses an end-to-end model based on a neural network. The end-to-end model based on a neural network is an LSTM-CRF model combined with an attention mechanism. On the basis of the syntactic analysis, the semantic role of each predicate in the sentence is accurately identified. The semantic roles include agent, patient, time, and place, and the semantic information of domain knowledge is further understood.

4. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S3, the knowledge association algorithm based on machine learning is calculated based on a graph neural network model based on deep learning, and the graph neural network model based on deep learning is GraphNeuralNetwork. Feature extraction and representation learning are performed on the financial domain knowledge sub-graph and the ecological domain knowledge sub-graph, and the similarity between the two graphs is calculated using a graph matching algorithm to discover potential knowledge associations. Then, based on an association mining algorithm that combines rules and statistics, domain-specific rules are first formulated. The domain-specific rules include similarity rules based on financial product functions and ecological species survival strategies. Then, through statistical analysis methods, the statistical analysis methods include co-occurrence analysis and association rule mining, knowledge association relationships that conform to the rules are mined in large-scale domain data, so as to achieve the inherent effective connection between the two domain knowledge sub-graphs.

5. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S5, the internal rate of return of the profitability indicator is accurately calculated through a cash flow discount model, taking into account the cash inflow and outflow of the financial product throughout its life cycle, and the net present value of the profitability indicator is discounted based on the market interest rate and the expected return of the product to reflect the current value of the product, thereby improving the accuracy of the quantitative evaluation model of the competition and complementary relationship; The credit risk of the risk indicator is assessed using a credit scoring model based on big data, taking into account the financial status, credit record, and business stability of small and micro enterprises. The market risk of the risk indicator is quantified using a value-at-risk (VaR) model or a conditional value-at-risk (CVaR) model to measure the maximum loss that a financial product may face at a certain confidence level. The liquidity risk of the liquidity indicator is evaluated by calculating the time required for capital realization and the capital turnover efficiency index. The capital realization speed of the liquidity indicator is quantified by analyzing factors such as the trading activity and market depth of financial products. The capital turnover period of the liquidity indicator is calculated based on the business model of small and micro enterprises and the repayment arrangement of financial products. The interest cost in the cost indicator is accurately calculated according to the interest rate pricing formula of the product, and the handling fee rate in the cost indicator is obtained by summarizing and counting various handling fee items.

6. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S5, the personalized demand characteristics of the small and micro enterprises are realized by conducting a detailed demand survey on the small and micro enterprises, and the demand survey includes but is not limited to questionnaire surveys, face-to-face interviews and data analysis methods to collect the attention of small and micro enterprises to various attributes of financial products at different development stages, different industry characteristics and different business objectives; The method of assigning corresponding weights to different demand dimensions of small and micro enterprises specifically involves using the analytic hierarchy process (AHP) to construct a hierarchical model, decomposing the financial demand targets of small and micro enterprises into multiple levels of criteria and indicators, determining the relative importance of elements at each level through pairwise comparison, and thus calculating the weight vectors of different demand dimensions, and adjusting and optimizing the weights in combination with the objective weighting method of the entropy weight method to ensure that the weights can reflect both the subjective demand preferences of small and micro enterprises and the objective information of the data.

7. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S6, the genetic algorithm (GA) dynamically adjusts the crossover probability and mutation probability parameters, and adopts an adaptive parameter adjustment strategy according to the change of the fitness function value during the search process. For example, when the fitness of the individuals in the population tends to be consistent, the mutation probability is appropriately increased to increase the diversity of the population and avoid the algorithm from falling into the local optimal solution. The particle swarm optimization algorithm (PSO) dynamically adjusts the inertia weight, learning factor and other parameters according to the flight speed and position changes of the particles, so that the particles achieve a balance between global search and local search, thereby improving the search efficiency and accuracy of the algorithm. The simulated annealing algorithm (SA) adaptively adjusts the cooling rate according to the changes in the current temperature and the objective function value to ensure that the algorithm converges to the global optimal solution or an approximate global optimal solution within a reasonable time.

8. The method for intelligently recommending financial technology products for small and micro enterprises according to claim 1, characterized in that: In step S7, the macroeconomic situation analysis is to select key macroeconomic indicators such as GDP growth rate, inflation rate and interest rate level, establish an econometric model between macroeconomic indicators and financial product demand and returns, and evaluate the market adaptability of the financial product information combination by analyzing the impact of changes in macroeconomic variables on the financial product market; The financial market fluctuations are analyzed by using a time series analysis model combined with a machine learning algorithm to analyze the main indicators of the financial market, comprehensively judge the future trend of the financial market, and provide a basis for adjusting the information combination of the financial products. The time series analysis model adopts an ARIMA model, and the main indicators of the financial market are stock indexes, bond yields, and exchange rates. The machine learning algorithm is a support vector machine (SVM) algorithm; The industry development trend analysis analyzes the market concentration, competitive landscape, and threat factors of new entrants in the industry in which small and micro enterprises are located, and uses Porter's Five Forces Model tool to evaluate the impact of the industry's competitive environment on the demand for financial products, ensuring that the recommended financial product information portfolio can adapt to the dynamic changes in industry development.

Images