A financial product recommendation strategy generation method, device, equipment, medium and product

Abstract

The application discloses a financial product recommendation strategy generation method, device, equipment, medium and product, and is applied to the financial field. The method comprises the following steps: acquiring user behavior data of a target user and user financial data of a user type to which the target user belongs; constructing a user behavior feature set according to the user behavior data; constructing a user data feature set according to the user financial data; determining product feature data of candidate financial products associated with a product query request; determining product matching degrees between the target user and each candidate financial product based on a product demand prediction model according to the user behavior feature set and the product feature data; determining credit risk values of each financial product for the target user based on a user risk prediction model according to the user behavior feature set, the user data feature set and the product feature data; and generating a financial product recommendation strategy according to the product matching degrees and the credit risk values, so that the target user determines a target financial product according to the financial product recommendation strategy.

Description

Technical Field

[0001] This invention relates to the field of financial technology, and in particular to a method, apparatus, device, medium, and product for generating financial product recommendation strategies. Background Technology

[0002] In recent years, with the improvement of national consumption levels and the change of consumption concepts, the market size of financial products has grown rapidly. However, facing the diversified market competition, traditional banks urgently need to adjust their business development strategies, leverage new technologies to fully utilize their own advantages, and achieve distinctive and personalized recommendations in order to seize market share in various financial products and better manage their financial product business.

[0003] In existing technologies, financial product recommendation schemes primarily rely on static scoring models. These models generate fixed credit scores for users by acquiring their historical credit data, and then filter target customers for various financial products based on these scores. Subsequently, the same type of financial product is pushed to all target users, with only basic discounts adjusted manually. However, static models cannot capture real-time changes in target users' needs, and recommending the same type of financial product to all target users results in extremely low matching between target users and recommended financial products, leading to very low recommendation efficiency and insufficient accuracy in risk assessment. Furthermore, separating product recommendation from credit risk assessment—through a second review of potential customers followed by separate credit risk assessment—also significantly impacts the efficiency of financial product recommendation.

[0004] Therefore, how to achieve personalized recommendations for financial products, improve the comprehensiveness and reliability of product risk assessment, and enhance the efficiency and accuracy of financial product recommendations has become an urgent technical problem to be solved. Summary of the Invention

[0005] This invention provides a method, apparatus, device, medium, and product for generating financial product recommendation strategies, so as to realize personalized recommendations for financial products, improve the comprehensiveness and reliability of product risk assessment, and enhance the efficiency and accuracy of financial product recommendations.

[0006] According to one aspect of the present invention, a method for generating a financial product recommendation strategy is provided, comprising: In response to a target user's product query request for financial products, obtain the target user's user behavior data and the user financial data of the user type to which the target user belongs; Based on the user behavior data, construct a user behavior feature set for the target user, and based on the user financial data, construct a user data feature set for the user type to which the target user belongs; Identify at least one candidate financial product associated with the product query request, as well as product feature data for each candidate financial product; Based on the user behavior feature set and the product feature data, and using a pre-built product demand prediction model, the product matching degree between the target user and each of the candidate financial products is determined; and based on the user behavior feature set, the user data feature set, and the product feature data, and using a pre-built user risk prediction model, the credit risk value of the target user for each of the candidate financial products is determined. Based on the product matching degree and the credit risk value, a financial product recommendation strategy is generated for the target user, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0007] According to another aspect of the present invention, a financial product recommendation strategy generation apparatus is provided, comprising: The user data acquisition module is used to respond to a target user's product query request for financial products and acquire the target user's user behavior data and the user financial data of the user type to which the target user belongs. The feature set construction module is used to construct a user behavior feature set of the target user based on the user behavior data, and to construct a user data feature set of the user type to which the target user belongs based on the user financial data; The product data determination module is used to determine at least one candidate financial product associated with the product query request and the product feature data of each candidate financial product. The feature data processing module is used to determine the product matching degree between the target user and each of the candidate financial products based on the user behavior feature set and the product feature data and a pre-built product demand prediction model; and to determine the credit risk value of the target user for each of the candidate financial products based on the user behavior feature set, the user data feature set and the product feature data and a pre-built user risk prediction model. The recommendation strategy generation module is used to generate a financial product recommendation strategy for the target user based on the product matching degree and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0008] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: at least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the financial product recommendation strategy generation method according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the financial product recommendation strategy generation method according to any embodiment of the present invention.

[0010] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the financial product recommendation strategy generation method according to any embodiment of the present invention.

[0011] The technical solution of this invention, in response to a target user's product query request for a financial product, acquires the target user's user behavior data and the user's financial data of the user type to which the target user belongs; constructs a user behavior feature set for the target user based on the user behavior data, and constructs a user data feature set for the user type to which the target user belongs based on the user financial data; determines at least one candidate financial product associated with the product query request and the product feature data of each candidate financial product; determines the product matching degree between the target user and each candidate financial product based on the user behavior feature set and the product feature data, using a pre-built product demand prediction model; and determines the credit risk value of the target user for each candidate financial product based on the user behavior feature set, the user data feature set, and the product feature data, using a pre-built user risk prediction model; and generates a financial product recommendation strategy for the target user based on the product matching degree and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy. This technical solution can determine the product matching degree between the target user and each candidate financial product, and determine the credit risk value of the target user for each candidate financial product, based on user behavior data, user financial data, and product feature data. Furthermore, based on product matching degree and credit risk value, a financial product recommendation strategy is generated for the target user, enabling the target user to determine the target financial product according to the financial product recommendation strategy, realizing personalized recommendations for financial products, improving the comprehensiveness and reliability of product risk assessment, and improving the efficiency and accuracy of financial product recommendations.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of a financial product recommendation strategy generation method according to Embodiment 1 of the present invention; Figure 2 This is a flowchart of a financial product recommendation strategy generation method according to Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of a financial product recommendation strategy generation device according to Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the financial product recommendation strategy generation method of this invention. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0016] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] Example 1 Figure 1This is a flowchart illustrating a method for generating financial product recommendation strategies according to Embodiment 1 of the present invention. This embodiment is applicable to situations where, during the recommendation process of financial products, the generated product recommendation strategy lacks sufficient risk assessment for the user and has a low degree of matching with the user, resulting in low efficiency and poor accuracy in recommending financial products. This method can be executed by a financial product recommendation strategy generation device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes: S110. In response to a target user's product query request for financial products, obtain the target user's user behavior data and the user financial data of the user type to which the target user belongs.

[0018] User behavior data can be data describing the behavior of target users in the process of using financial products and services. Specifically, it can include financial transaction data, interaction behavior data, credit behavior data, and basic attribute data. User financial transaction data can include transfer amount, transaction frequency, investment holdings, and credit repayment records. Interaction behavior data can include product click count, product page visit duration, product consultation records, and product page dwell time. Credit behavior data can include historical credit count, credit limit data ratio, repayment method preference, and historical product default records. Basic attribute data can include age, occupation, income level, credit rating, and risk preference.

[0019] The user financial data can be statistical analysis of the behavioral data of a user group to which they belong, reflecting the financial behavior of that group. Specifically, this can include historical default rates, product default distribution, average credit utilization rate, average loan application frequency, and credit volatility coefficient. The user type can be a homogeneous grouping of different users achieved through clustering algorithms, ensuring that users in the same group are highly similar in financial behavior characteristics, credit level, and risk attributes. Alternatively, clustering can be formed using at least one dimension: age, occupation, income level, and credit rating.

[0020] Specifically, based on the target's product query request for financial products, such as when a target user clicks on the product details page of a financial product (which may include credit products, wealth management products, and insurance products) to query and inquire about the product details of the relevant financial products, user behavior data of the target user can be obtained. Based on the user behavior data, the user type of the target user can be determined, and based on the user type of the target user, user financial data of the user type of the target user can be obtained.

[0021] S120. Based on user behavior data, construct a user behavior feature set for the target user, and based on user financial data, construct a user data feature set for the user type to which the target user belongs.

[0022] The user behavior feature set can be a quantitative dataset obtained by structurally analyzing the behavioral data of target users. Specifically, it can include credit behavior features, financial management behavior features, basic behavior features, and credit behavior features. Credit behavior features can include credit application frequency, credit application type preference, and credit usage amount; financial management behavior features can include product type preference, risk preference type, and investment frequency; basic behavior features can include account active days, fund retention rate, and average monthly deposit amount; and credit behavior features can include timely repayment rate, credit inquiry frequency, credit score, and minimum repayment frequency. Specifically, features can be extracted from the target user's behavioral data to obtain the target user's behavioral feature data. Depending on the data type (numerical or categorical), a user behavior feature set for the target user can be constructed based on a pre-defined feature clustering algorithm.

[0023] The user data feature set can be a group-based quantitative dataset obtained by statistically analyzing the individual behavioral data of all users within the target user's user type. It can also be a comprehensive data set including the target user and similar users' basic attributes, credit records, and behavioral patterns. Specifically, it can include asset-liability characteristics, credit performance characteristics, financial behavior characteristics, and risk attribute characteristics. Asset-liability characteristics can include average monthly income, average debt ratio, average monthly mortgage payment pressure, and the percentage of homeowners. Credit performance characteristics can include average credit score, credit score volatility coefficient, the percentage of users with no credit history, and the percentage of users with high credit scores. Financial behavior characteristics can include average monthly product inquiry frequency, credit product preference percentage, average credit utilization rate, and loan approval rate. Risk attribute characteristics can include historical default rate, average timely repayment rate, the percentage of users with severe delinquencies, and the product default distribution ratio. In essence, based on user financial data, the financial data characteristics of the target user's user type can be determined, and a user data feature set for that user type can be constructed based on these financial data characteristics.

[0024] S130. Determine at least one candidate financial product associated with the product query request and the product feature data of each candidate financial product.

[0025] The product characteristic data can be quantitative data describing the product attributes of a financial product, specifically quantitative data reflecting the product's own risk attributes. This includes historical default rates, non-performing loan rates, risk levels, and entry thresholds, where entry thresholds may include minimum credit scores and maximum debt ratios. Specifically, based on the target user's product query request, the product type of the financial product requested by the target user and at least one candidate financial product associated with that product type can be determined, along with the risk attribute data of each candidate financial product. For example, if the target user's query type is a credit product, then associated candidate financial products such as credit loans, consumer loans, and mortgage loans can be determined. Furthermore, the product characteristic data of each candidate financial product can be determined, such as a historical default rate of 5%, a non-performing loan rate of 3%, a low to medium risk level, and entry thresholds of a good credit score and a debt ratio not exceeding 40%. This embodiment does not impose specific limitations on these.

[0026] S140. Based on the user behavior feature set and product feature data, and using a pre-built product demand prediction model, determine the product matching degree between the target user and each candidate financial product; and based on the user behavior feature set, user data feature set, and product feature data, and using a pre-built user risk prediction model, determine the credit risk value of the target user for each candidate financial product.

[0027] Among them, product matching degree refers to the degree of fit between the product attribute information of financial products and the user attribute information of target users. Specifically, the user behavior feature set of target users and the product feature data of each candidate financial product can be input into a pre-built product demand prediction model to obtain the feature matching degree between each dimension of the user behavior feature set and the product feature data, and the product matching degree between the target user and each candidate financial product can be determined based on the feature matching degree. For example, if it can be determined that the candidate financial products associated with the target user are credit products, including products A, B, and C, and further, the user behavior characteristics set of the target user can be determined to include product type preference (credit type), credit application type preference (low-risk type), excellent credit score, personal asset-liability ratio (3%), and average monthly income of 15,000 yuan, where the product characteristic data of product A can include product risk level (low risk), product entry threshold (good credit risk level), and average monthly income of 8,000 yuan; the product characteristic data of product B can include product risk level (medium risk), product entry threshold (good credit risk level), and average monthly income of 10,000 yuan; and the product characteristic data of product C can include product risk level (low risk), product entry threshold (average credit risk level), and average monthly income of 5,000 yuan. Then it can be determined that the product matching degree of the target user with product A is 95%, with product B is 30%, and with product C is 75%.

[0028] The credit risk value can be a quantitative calculation of the credit risk faced by a financial product. It can be used to reflect the probability of loss caused by credit events such as default by the target user or downgrade of credit rating within a future period. Specifically, it can refer to the probability of default by the target user on each candidate product. Specifically, user behavior data, user feature datasets, and product feature data can be input into a pre-built user risk prediction model. Based on the user feature dataset and product feature data, the product default probability of the target user's user type on each candidate financial product is obtained. The credit risk value of the target user on each candidate financial product is then determined based on the product default probability. For example, continuing the previous example, based on the user feature dataset and product feature data, if the probability of default for the target user's user type on product A is determined to be 40%, on product B to be 10%, and on product C to be 20%, then the credit risk value for the target user on product A is determined to be 0.4, on product B to be 0.1, and on product C to be 0.2. This embodiment does not impose specific limitations on this.

[0029] Optionally, based on user behavior feature sets and product feature data, and using a pre-built product demand prediction model, the product matching degree between the target user and each candidate financial product is determined, including: determining the target user's behavioral feature data in at least one feature dimension based on the user behavior feature set; obtaining the feature similarity between the target user and each candidate financial product in multiple feature dimensions based on the behavioral feature data and product feature data, and using a pre-built product demand prediction model; and determining the product matching degree between the target user and each candidate financial product based on the feature similarity.

[0030] Behavioral feature data refers to various behavioral data of target users towards financial products, specifically including investment behavior data, risk preference data, and basic attribute data. Specifically, based on the user behavior feature set, behavioral feature data of target users across various dimensions can be extracted to obtain the target user's behavioral feature vector. This could include a risk level of "conservative," an investment horizon of medium to long term, and a minimum investment amount preference of less than 50,000 yuan. Based on the product feature data, product feature vectors for each candidate financial product are determined, establishing a mapping relationship between the target user's behavioral feature vector and the product feature vectors of each candidate financial product. For example, the user's risk preference dimension corresponds to the product's risk level dimension, and the user's investment horizon preference corresponds to the product's investment horizon dimension. These mapped behavioral feature vectors and product feature vectors are then input into a pre-built product demand prediction model to obtain the feature similarity between the target user's behavioral feature vector and the product feature vectors of each candidate financial product across various feature dimensions. Furthermore, the feature weight values ​​of each behavioral characteristic data can be determined. Based on the feature weight values ​​and feature similarity, the product matching degree between the target user and each candidate financial product can be determined. For example, the feature weight value corresponding to the risk preference dimension is 0.4, the feature weight value corresponding to the investment period dimension is 0.3, and the feature weight value corresponding to the minimum investment amount dimension is 0.3. Further, if the product characteristic data of a candidate financial product are: low risk level, short-to-medium term investment, and minimum investment amount of 30,000 yuan, then the feature similarity corresponding to the risk preference dimension can be determined to be 1, the feature similarity corresponding to the investment period dimension to be 0.8, and the feature similarity corresponding to the minimum investment amount dimension to be 0.6. Therefore, the product matching degree between the target user and the candidate financial product can be determined to be 82%. This embodiment does not impose specific limitations on this.

[0031] This technical solution can determine the feature similarity between the target user and each candidate financial product across multiple feature dimensions based on user behavior feature sets and product feature data, thereby obtaining the product matching degree between the target user and each candidate financial product, realizing personalized recommendations for financial products, and improving the efficiency of financial product recommendations.

[0032] Optionally, based on user behavior feature sets, user data feature sets, and product feature data, and using a pre-built user risk prediction model, the credit risk value of the target user for each candidate financial product is determined, including: determining the user risk feature parameters of the target user based on the user behavior feature set; determining the historical credit risk value of the user type to which the target user belongs based on the user data feature set; determining the risk feature parameters of each candidate financial product based on the product feature data; and determining the credit risk value of the target user for each candidate financial product based on the risk feature parameters, historical credit risk value, and risk feature parameters, using a pre-built user risk prediction model.

[0033] User risk characteristic parameters can be quantitative indicators used to assess a user's potential risk. These can include loan delinquency frequency, user debt ratio, and credit score, reflecting risk attributes such as the target user's risk preference, risk tolerance, and risk behavior habits. Specifically, risk behavior data related to the target user can be obtained by filtering user behavior feature sets. Based on this data, user risk characteristic parameters can be determined, including parameters such as loan delinquency rate, high-risk investment ratio, fund fluctuation, and asset-liability ratio. Furthermore, based on the user data feature set, historical credit data corresponding to the target user's user type can be determined, and the historical credit risk value for that user type can be calculated.

[0034] Risk characteristic parameters can be index parameters used to assess the product risk of financial products. Specifically, they can include market risk parameters, credit risk parameters, and interest rate risk parameters, which can reflect the product's inherent risk level, risk triggering conditions, credit correlation requirements, etc. Specifically, based on product characteristic data, the corresponding product risk parameters for each candidate financial product can be determined, such as product risk level, credit requirements, and historical default data. Based on these product risk parameters, the risk characteristic parameters for each candidate financial product can be calculated. Then, the risk characteristic parameters, historical credit risk values, and risk characteristic parameters are input into a pre-built user risk prediction model to obtain the target user's product risk value for each candidate financial product, as output by the model.

[0035] This technical solution can determine the credit risk value of a target user for each candidate financial product based on the user's risk characteristic parameters, historical credit risk value, and risk characteristic parameters, thereby improving the efficiency and accuracy of risk assessment of financial products and enabling personalized recommendations for financial products.

[0036] S150. Based on product matching degree and credit risk value, generate a financial product recommendation strategy for the target user, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0037] Specifically, based on product matching degree and credit risk value, at least one financial product matching the target user can be selected from each candidate financial product. The product information of each financial product to be recommended can be determined, and the financial products to be recommended can be ranked. Then, based on the product ranking results and product information of each financial product to be recommended, a financial product recommendation strategy for the target user can be generated, so that the target user can select the target financial product according to the financial product recommendation strategy.

[0038] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant regions.

[0039] The technical solution of this invention, in response to a target user's product query request for a financial product, acquires the target user's user behavior data and the user's financial data of their respective user type; constructs a user behavior feature set for the target user based on the user behavior data, and constructs a user data feature set for the target user's user type based on the user financial data; determines at least one candidate financial product associated with the product query request and the product feature data of each candidate financial product; determines the product matching degree between the target user and each candidate financial product based on the user behavior feature set and the product feature data, using a pre-built product demand prediction model; and determines the target user's credit risk value for each financial product based on the user behavior feature set, the user data feature set, and the product feature data, using a pre-built user risk prediction model; and generates a financial product recommendation strategy for the target user based on the product matching degree and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy. This technical solution can determine the product matching degree between the target user and each candidate financial product, and determine the target user's credit risk value for each financial product, based on user behavior data, user financial data, and product feature data. Furthermore, based on product matching degree and credit risk value, a financial product recommendation strategy is generated for the target user, enabling the target user to determine the target financial product according to the financial product recommendation strategy, realizing personalized recommendations for financial products, improving the comprehensiveness and reliability of product risk assessment, and improving the efficiency and accuracy of financial product recommendations.

[0040] Example 2 Figure 2 This is a flowchart of a financial product recommendation strategy generation method provided in Embodiment 2 of the present invention. This embodiment further optimizes the above-mentioned financial product recommendation strategy generation method based on the embodiments described above.

[0041] Furthermore, the step "generating a financial product recommendation strategy for the target user based on product matching degree and credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy" is refined to "determining whether the target user meets the pre-set anomaly identification conditions based on product matching degree; if so, performing a consistency check on the credit risk value according to the pre-set credit risk verification rules to obtain the consistency verification result; determining the target user's abnormal risk level based on the consistency verification result; and generating a financial product recommendation strategy for the target user based on the abnormal risk level, product matching degree, and credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy." This improves the generation method of the financial product recommendation strategy. Figure 2 As shown, the method includes: S210. In response to a target user's product query request for financial products, obtain the target user's user behavior data and the user financial data of the user type to which the target user belongs.

[0042] S220. Based on user behavior data, construct a user behavior feature set for the target user, and based on user financial data, construct a user data feature set for the user type to which the target user belongs.

[0043] S230. Determine at least one candidate financial product associated with the product query request and the product feature data of each candidate financial product.

[0044] S240. Based on the user behavior feature set and product feature data, and using a pre-built product demand prediction model, determine the product matching degree between the target user and each candidate financial product; and based on the user behavior feature set, user data feature set, and product feature data, and using a pre-built user risk prediction model, determine the credit risk value of the target user for each financial product.

[0045] S250. Based on the product matching degree, determine whether the target user meets the pre-set matching degree abnormality identification conditions.

[0046] Specifically, determining whether a target user meets the pre-set anomaly identification criteria can be achieved by comparing the product matching degree with a pre-set product matching degree threshold. For example, the pre-set product matching degree threshold could be 75%. If the product matching degree between the target user and the candidate financial product is determined to be 80%, then the target user meets the pre-set anomaly identification criteria. If the product matching degree between the target user and the candidate financial product is determined to be 65%, then the target user does not meet the pre-set anomaly identification criteria.

[0047] S260. If so, then according to the pre-set credit risk verification rules, the credit risk value is verified for consistency, and the consistency verification result is obtained.

[0048] Specifically, the credit risk value of each candidate financial product by the target user can be verified against a pre-set credit risk range. For example, the pre-set credit risk range can be low risk level less than or equal to 0.2, medium risk level 0.2-0.4, and high risk level greater than 0.4. The credit risk value can then be matched with the pre-set credit risk range, and the credit risk range that matches the credit risk value can be determined as the consistency verification result of the credit risk value.

[0049] S270. Based on the consistency verification results, determine the abnormal risk level of the target user.

[0050] The abnormal risk level can be a quantitative indicator parameter of the severity of the potential risk of the target user. Specifically, it can be achieved by matching the credit risk value with a pre-set credit risk range to realize the consistency verification of credit risk, and determine the abnormal risk level of the target user based on the consistency verification result of credit risk. For example, if the credit risk value of the target user is 0.3, then the credit risk range matched by the credit risk value of the target user can be determined to be 0.2-0.4, and thus the abnormal risk level of the target user can be determined to be medium risk level. This embodiment does not impose specific limitations on this.

[0051] S280. Based on the abnormal risk level, product matching degree, and credit risk value, generate a financial product recommendation strategy for the target user, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0052] Specifically, based on the abnormal risk level, it can be determined whether the target user meets the pre-set product recommendation criteria. If so, then based on product matching degree and credit risk value, a financial product to be recommended that matches the target user can be determined, and a user product recommendation strategy can be generated based on the product information of the products to be recommended. Determining whether the target user meets the pre-set product recommendation criteria can be done by determining whether the target user's abnormal risk level is low or medium risk. Furthermore, if the target user's abnormal risk level is determined to be low risk, then the target user can be determined to meet the pre-set product recommendation criteria. Then, based on product matching degree and credit risk value, products to be recommended can be screened from various candidate financial products, and a product recommendation strategy can be generated for the recommended financial products, allowing the target user to select the target financial product according to the product recommendation strategy.

[0053] Optionally, a financial product recommendation strategy for the target user is generated based on the abnormal risk level, product matching degree, and credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy. This includes: determining whether the abnormal risk level meets the pre-set risk level identification conditions; if so, determining a first reference financial product based on the product matching degree and a pre-set matching degree threshold; and determining a second reference financial product based on the credit risk value and a pre-set risk value range; determining a first financial product to be recommended to the target user based on the first and second reference financial products, and generating a financial product recommendation strategy for the target user based on the first financial product to be recommended, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0054] Specifically, determining whether the abnormal risk level meets the pre-set risk level identification conditions can involve determining whether the target user's abnormal risk level is low or medium risk. Further, if the target user's abnormal risk level meets the pre-set risk level identification conditions, a first reference financial product and a second reference financial product can be determined based on product matching degree and credit risk value. Based on these two reference financial products, a financial product to be recommended to the target user can be determined. Finally, a product recommendation strategy for the target user is generated based on this strategy, and the target user selects the target financial product according to the recommendation strategy.

[0055] Optionally, a first reference financial product and a second reference financial product can be determined based on product matching degree and credit risk value, respectively. Specifically, the product matching degree can be compared with a pre-set matching degree threshold, and the product with the matching degree comparison result can be determined as the first reference financial product, where the matching degree threshold can be 0.7. Specifically, if the target user's product matching degree with a candidate financial product is greater than 0.7, then that candidate financial product can be determined as the first reference financial product. For example, the candidate financial products may include financial product 1, financial product 2, financial product 3, financial product 4, and financial product 5, where the target user's product matching degree with financial product 1 is 0.75, with financial products 2 and 3 it is 0.8, with financial product 4 it is 0.5, and with financial product 5 it is 0.4. Based on this, the first reference financial products can be determined as financial product 1, financial product 2, and financial product 3.

[0056] Furthermore, the credit risk values ​​of the target user for each candidate financial product can be matched with a pre-defined risk value range, and a second reference financial product can be determined based on the risk value matching results. The credit risk value range can be less than or equal to 0.4. For example, continuing the previous example, if the target user's credit risk value for financial product 1 is determined to be 0.45, for financial product 2 to be 0.15, and for financial product 3 to be 0.2, then financial products 2 and 3 can be determined as the second reference financial products. Further, financial products 2 and 3 can be identified as the financial products to be recommended, and a product recommendation strategy can be generated, allowing the target user to choose financial product 2 or financial product 3 as the target financial product.

[0057] This technical solution can identify abnormal risks of target users based on the level of abnormal risk, and then determine the financial products to be recommended to the target users based on product matching degree and credit risk value, and generate a financial product recommendation strategy for the target users, thereby realizing personalized recommendations of financial products and improving the recommendation strategy.

[0058] Optionally, after determining the first recommended financial product for the target user based on the first and second reference financial products, and generating a financial product recommendation strategy for the target user based on the first recommended financial product, so that the target user can determine the target financial product according to the financial product recommendation strategy, the method further includes: determining the target user's product response data to the candidate financial products in the financial product recommendation strategy; generating behavioral sequence characteristics for the target user based on the product response data; adjusting the model parameters of the product demand prediction model and the user risk prediction model based on the behavioral sequence characteristics to obtain the adjusted product demand prediction model and the adjusted user risk prediction model; determining the second recommended financial product for the target user based on the user behavior feature set, user data feature set, and product feature data, using the adjusted product demand prediction model and the adjusted user risk prediction model; and generating an optimized financial product recommendation strategy for the target user based on the second recommended financial product, so that the target user can determine the target financial product according to the optimized financial product recommendation strategy.

[0059] The product response data can include the number of times target users browse candidate financial products in the financial product recommendation strategy, the purchase ratio, and the default ratio. The number of browsing times refers to the number of times target users click to view the product details of candidate financial products in the financial product recommendation strategy. The purchase ratio is the ratio of the number of purchases to the number of clicks for candidate financial products in the financial product recommendation strategy. For example, if a target user clicks 10 times and purchases 5 times, the purchase ratio can be 50%. The default ratio is the percentage of times a target user fails to conduct product transactions according to the pre-set product transaction strategy after purchasing a candidate financial product. This can include the number of times a loan product is not repaid on the agreed repayment date and the number of times a wealth management product is sold before the holding period is met.

[0060] Furthermore, based on the response data, the user behavior sequence characteristics of the target user can be determined. Based on these characteristics, the product matching degree and credit risk value of the target user for each candidate financial product can be updated. Then, based on the updated product matching degree and credit risk value, the model parameters of the product demand prediction model and the user risk prediction model can be adjusted, resulting in the adjusted product demand prediction model and the adjusted user risk prediction model. Next, the user behavior feature set, user data feature set, and product feature data are respectively input into the adjusted product demand prediction model and the adjusted user risk prediction model to determine the second financial product to be recommended to the target user. Based on this second financial product, the financial product recommendation strategy for the target user can be optimized, enabling the target user to select the target financial product according to the optimized strategy.

[0061] This technical solution can construct behavioral sequence characteristics of target users based on their product response data to candidate financial products in the financial product recommendation strategy. Based on this, the product demand prediction model and user risk prediction model can be adjusted to optimize the financial product recommendation strategy for target users. This achieves personalized recommendations for financial products, improves the efficiency and accuracy of financial product recommendations, and enhances the accuracy of product risk assessment for financial products.

[0062] The technical solution of this invention determines whether a target user meets pre-set anomaly identification conditions based on product matching degree. If so, it performs a consistency check on the credit risk value according to pre-set credit risk verification rules to obtain a consistency verification result. Based on the consistency verification result, it determines the target user's abnormal risk level. Based on the abnormal risk level, product matching degree, and credit risk value, it generates a financial product recommendation strategy for the target user, enabling the target user to select a target financial product according to the recommendation strategy. This technical solution can identify the product matching degree between the target user and each candidate financial product, and perform a consistency check on the credit risk value according to credit risk verification rules to determine the target user's abnormal risk level, thereby generating a financial product recommendation strategy for the target user. This achieves personalized recommendations for financial products and, by identifying the target user's abnormal risk level, greatly improves the comprehensiveness and reliability of financial product risk assessment, and enhances the efficiency and accuracy of financial product recommendations.

[0063] Example 3 Figure 3 This is a schematic diagram of a financial product recommendation strategy generation device provided in Embodiment 3 of the present invention. The financial product recommendation strategy generation device provided in this embodiment of the present invention is applicable to situations involving multi-level, multi-role user structures in power Internet of Things systems. This financial product recommendation strategy generation device can be implemented in hardware and / or software, such as... Figure 3 As shown, it specifically includes: a user data acquisition module 310, a feature set construction module 320, a product data determination module 330, a feature data processing module 340, and a recommendation strategy generation module 350. Among them, User data acquisition module 310 is used to respond to a target user's product query request for financial products and acquire the target user's user behavior data and the user financial data of the user type to which the target user belongs; The feature set construction module 320 is used to construct a user behavior feature set of the target user based on the user behavior data, and to construct a user data feature set of the user type to which the target user belongs based on the user financial data; Product data determination module 330 is used to determine at least one candidate financial product associated with the product query request and product feature data of each candidate financial product; The feature data processing module 340 is used to determine the product matching degree between the target user and each of the candidate financial products based on the user behavior feature set and the product feature data and a pre-built product demand prediction model; and to determine the credit risk value of the target user for each of the candidate financial products based on the user behavior feature set, the user data feature set and the product feature data and a pre-built user risk prediction model. The recommendation strategy generation module 350 is used to generate a financial product recommendation strategy for the target user based on the product matching degree and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0064] This solution can determine the product matching degree between a target user and each candidate financial product, as well as the credit risk value of the target user for each financial product, based on user behavior data, user financial data, and product feature data. Then, based on the product matching degree and credit risk value, a financial product recommendation strategy is generated for the target user. This allows the target user to select target financial products according to the recommendation strategy, achieving personalized recommendations, improving the comprehensiveness and reliability of product risk assessment, and increasing the efficiency and accuracy of financial product recommendations.

[0065] Optionally, the feature data processing module 340 is specifically used to determine the behavioral feature data of the target user in at least one feature dimension based on the user behavior feature set. Based on the behavioral feature data and the product feature data, and using a pre-built product demand prediction model, the feature similarity between the target user and each of the candidate financial products is obtained across multiple feature dimensions. Based on the feature similarity, the product matching degree between the target user and each of the candidate financial products is determined.

[0066] Optionally, the feature data processing module 340 is specifically used to determine the user risk feature parameters of the target user based on the user behavior feature set; and to determine the historical credit risk value of the user type to which the target user belongs based on the user data feature set. Based on the product feature data, the risk characteristic parameters of each candidate financial product are determined; Based on the risk characteristic parameters, the historical credit risk value, and the risk characteristic parameters, and using a pre-built user risk prediction model, the credit risk value of the target user for each of the financial products is determined.

[0067] Optionally, the recommendation strategy generation module 350 is specifically used to determine whether the target user meets the pre-set abnormal matching conditions based on the product matching degree. If so, then according to the pre-set credit risk verification rules, the credit risk value is verified for consistency, and the consistency verification result is obtained. Based on the consistency verification results, the abnormal risk level of the target user is determined; Based on the abnormal risk level, the product matching degree, and the credit risk value, a financial product recommendation strategy is generated for the target user, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0068] Optionally, the recommended strategy generation module 350 is further used to determine whether the abnormal risk level meets the preset risk level identification conditions; If so, a first reference financial product is determined based on the product matching degree and a pre-set matching degree threshold; and a second reference financial product is determined based on the credit risk value and a pre-set risk value range. Based on the first reference financial product and the second reference financial product, a first financial product to be recommended to the target user is determined, and a financial product recommendation strategy for the target user is generated based on the first financial product to be recommended, so that the target user can determine the target financial product according to the financial product recommendation strategy.

[0069] Optionally, the device may also include: The product recommendation strategy optimization module is used to determine the first financial product to be recommended to the target user based on the first reference financial product and the second reference financial product, and to generate a financial product recommendation strategy for the target user based on the first financial product to be recommended, so that the target user determines the target financial product according to the financial product recommendation strategy, and then determines the product response data of the target user to the candidate financial products in the financial product recommendation strategy. Based on the product response data, generate behavioral sequence features for the target user; Based on the behavioral sequence characteristics, the model parameters of the product demand forecasting model and the user risk forecasting model are adjusted to obtain the adjusted product demand forecasting model and the adjusted user risk forecasting model. Based on the user behavior feature set, the user data feature set, and the product feature data, and using the adjusted product demand prediction model and the adjusted user risk prediction model, a second financial product to be recommended for the target user is determined. Based on the second financial product to be recommended, an optimized financial product recommendation strategy is generated for the target user, so that the target user can determine the target financial product according to the optimized financial product recommendation strategy.

[0070] The financial product recommendation strategy generation device provided in this embodiment of the invention can execute the financial product recommendation strategy generation method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0071] Example 4 Figure 4 A schematic diagram of an electronic device 40 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0072] like Figure 4 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 and a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded from storage unit 48 into the RAM 43. The RAM 43 can also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0073] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0074] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as financial product recommendation strategy generation methods.

[0075] In some embodiments, the financial product recommendation strategy generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the financial product recommendation strategy generation method described above may be performed. Alternatively, in other embodiments, processor 41 may be configured to perform the financial product recommendation strategy generation method by any other suitable means (e.g., by means of firmware).

[0076] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0077] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0078] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0079] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0080] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0081] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability.

[0082] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0083] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for generating a financial product recommendation strategy, characterized in that, include: In response to a target user's product query request for financial products, obtain the target user's user behavior data and the user financial data of the user type to which the target user belongs; Based on the user behavior data, construct a user behavior feature set for the target user, and based on the user financial data, construct a user data feature set for the user type to which the target user belongs; Identify at least one candidate financial product associated with the product query request, as well as product feature data for each candidate financial product; Based on the user behavior feature set and the product feature data, and using a pre-built product demand prediction model, the product matching degree between the target user and each of the candidate financial products is determined; and based on the user behavior feature set, the user data feature set, and the product feature data, and using a pre-built user risk prediction model, the credit risk value of the target user for each of the candidate financial products is determined. Based on the product matching degree and the credit risk value, a financial product recommendation strategy is generated for the target user, so that the target user can determine the target financial product according to the financial product recommendation strategy.

2. The method according to claim 1, characterized in that, The step of determining the product matching degree between the target user and each of the candidate financial products based on the user behavior feature set and the product feature data, and using a pre-built product demand prediction model, includes: Based on the user behavior feature set, determine the behavioral feature data of the target user in at least one feature dimension; Based on the behavioral feature data and the product feature data, and using a pre-built product demand prediction model, the feature similarity between the target user and each of the candidate financial products is obtained across multiple feature dimensions. Based on the feature similarity, the product matching degree between the target user and each of the candidate financial products is determined.

3. The method according to claim 1, characterized in that, The step of determining the credit risk value of the target user for each of the candidate financial products based on the user behavior feature set, the user data feature set, and the product feature data, and on a pre-built user risk prediction model, includes: Based on the user behavior feature set, determine the user risk feature parameters of the target user; and based on the user data feature set, determine the historical credit risk value of the user type to which the target user belongs; Based on the product feature data, the risk characteristic parameters of each candidate financial product are determined; Based on the risk characteristic parameters, the historical credit risk value, and the risk characteristic parameters, and using a pre-built user risk prediction model, the credit risk value of the target user for each of the candidate financial products is determined.

4. The method according to claim 1, characterized in that, The step of generating a financial product recommendation strategy for the target user based on the product matching degree and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy, includes: Based on the product matching degree, determine whether the target user meets the pre-set matching degree anomaly identification conditions; If so, then according to the pre-set credit risk verification rules, the credit risk value is verified for consistency, and the consistency verification result is obtained. Based on the consistency verification results, the abnormal risk level of the target user is determined; Based on the abnormal risk level, the product matching degree, and the credit risk value, a financial product recommendation strategy is generated for the target user, so that the target user can determine the target financial product according to the financial product recommendation strategy.

5. The method according to claim 4, characterized in that, The step of generating a financial product recommendation strategy for the target user based on the abnormal risk level, the product matching degree, and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy, includes: Determine whether the abnormal risk level meets the pre-set risk level identification conditions; If so, a first reference financial product is determined based on the product matching degree and a pre-set matching degree threshold; and a second reference financial product is determined based on the credit risk value and a pre-set risk value range. Based on the first reference financial product and the second reference financial product, a first financial product to be recommended to the target user is determined, and a financial product recommendation strategy for the target user is generated based on the first financial product to be recommended, so that the target user can determine the target financial product according to the financial product recommendation strategy.

6. The method according to claim 5, characterized in that, After determining a first recommended financial product for the target user based on the first reference financial product and the second reference financial product, and generating a financial product recommendation strategy for the target user based on the first recommended financial product, so that the target user determines the target financial product according to the financial product recommendation strategy, the method further includes: Determine the target user's product response data to candidate financial products in the financial product recommendation strategy; Based on the product response data, generate behavioral sequence features for the target user; Based on the behavioral sequence characteristics, the model parameters of the product demand forecasting model and the user risk forecasting model are adjusted to obtain the adjusted product demand forecasting model and the adjusted user risk forecasting model. Based on the user behavior feature set, the user data feature set, and the product feature data, and using the adjusted product demand prediction model and the adjusted user risk prediction model, a second financial product to be recommended for the target user is determined. Based on the second financial product to be recommended, an optimized financial product recommendation strategy is generated for the target user, so that the target user can determine the target financial product according to the optimized financial product recommendation strategy.

7. A financial product recommendation strategy generation device, characterized in that, include: The user data acquisition module is used to respond to a target user's product query request for financial products and acquire the target user's user behavior data and the user financial data of the user type to which the target user belongs. The feature set construction module is used to construct a user behavior feature set of the target user based on the user behavior data, and to construct a user data feature set of the user type to which the target user belongs based on the user financial data; The product data determination module is used to determine at least one candidate financial product associated with the product query request and the product feature data of each candidate financial product. The feature data processing module is used to determine the product matching degree between the target user and each of the candidate financial products based on the user behavior feature set and the product feature data and a pre-built product demand prediction model; and to determine the credit risk value of the target user for each of the candidate financial products based on the user behavior feature set, the user data feature set and the product feature data and a pre-built user risk prediction model. The recommendation strategy generation module is used to generate a financial product recommendation strategy for the target user based on the product matching degree and the credit risk value, so that the target user can determine the target financial product according to the financial product recommendation strategy.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that is executed by the at least one processor, such that the at least one processor is able to perform the financial product recommendation strategy generation method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the financial product recommendation strategy generation method according to any one of claims 1-6.

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

Images