Financial product recommendation method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a financial product recommendation method and device, electronic equipment and a storage medium, and belongs to the technical field of artificial intelligence. The method comprises the steps of obtaining a historical transaction record of a target user, predicting an optimal cash holding amount of the target user according to the historical transaction record, and determining a plurality of preference financial products of the target user according to the historical transaction record; querying a plurality of preset target financial products, and determining the product similarity between each target financial product and the preference financial product; screening a plurality of to-be-recommended financial products from the target financial products according to the product similarity; and determining a financial product combination according to the optimal cash holding amount and the to-be-recommended financial product, determining an expected income and an investment style of the financial product combination, generating a financial product recommendation scheme according to the financial product combination, the expected income and the investment style, and recommending the financial product recommendation scheme to the target user. According to the embodiment of the invention, the accuracy of financial product recommendation and the user stickiness can be improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method and apparatus for recommending financial products, an electronic device, and a storage medium. Background Technology

[0002] As competition intensifies and financial innovation deepens, users are facing increasingly diverse and complex choices of financial products, leading to a rise in the time cost for users to understand various financial products. Without a thorough understanding of different financial products, users instinctively prefer to choose those they are more familiar with for asset appreciation, resulting in high concentration of users on a single type of product and low user stickiness.

[0003] In the industry, research on product portfolio management mainly focuses on the following two aspects:

[0004] 1) Portfolio Strategy: By analyzing overall market trends, formulate a series of effective product portfolio strategies and promote them.

[0005] 2) Product substitution: By analyzing each product in the product portfolio, identify products with problems and declining sales, provide alternative solutions, and optimize the product structure.

[0006] Currently, product portfolio recommendations are mainly made through recommendations from account managers and discussions with users. In the process of product portfolio management, on the one hand, it is impossible to guarantee that all account managers are thoroughly familiar with all products, and they are prone to having fixed ideas when making product portfolios, resulting in unsatisfactory portfolio effects, affecting the accuracy of financial product recommendations, and failing to meet user needs; on the other hand, due to the lack of relevant data support, it is difficult to convince users based solely on experience, which fails to increase user trust and thus affects user stickiness. Summary of the Invention

[0007] The main objective of this application is to provide a financial product recommendation method, apparatus, electronic device, and storage medium, aiming to improve the accuracy of financial product recommendations and user stickiness.

[0008] To achieve the above objectives, a first aspect of this application proposes a method for recommending financial products, the method comprising:

[0009] Obtain the target user's historical transaction records, predict the target user's optimal cash holding amount based on the historical transaction records, and determine several preferred financial products of the target user based on the historical transaction records;

[0010] Query multiple preset target financial products and determine the product similarity between each target financial product and the preferred financial product;

[0011] Based on the product similarity, multiple financial products to be recommended are selected from the target financial products;

[0012] A financial product portfolio is determined based on the optimal cash holding and the financial products to be recommended, and the expected return and investment style of the financial product portfolio are determined. Then, a financial product recommendation plan is generated based on the financial product portfolio, the expected return, and the investment style.

[0013] In some embodiments, predicting the optimal cash holding amount for the target user based on the historical transaction records includes:

[0014] Based on the historical transaction records, determine the target user's current securities information and the first total cash requirement in each preset historical period;

[0015] Predict the second total cash demand of the target user in the target period based on the first total cash demand.

[0016] Determine the target user's transaction costs and opportunity cost rate based on the current securities information;

[0017] The optimal cash holding is determined based on the second total cash demand, the transaction costs, and the opportunity cost rate.

[0018] In some embodiments, determining the target user's preferred financial products based on the historical transaction records includes:

[0019] Based on the historical transaction records, determine the historical purchase frequency, historical purchase volume, and historical holding duration of multiple historical financial products;

[0020] The user preference score for each historical financial product is obtained by weighting and summing the historical purchase frequency, historical purchase share, and historical holding period according to the preset weighting coefficients.

[0021] Based on the user preference score, select a number of preferred financial products from the historical financial products.

[0022] In some embodiments, determining the product similarity between each of the target financial products and the preferred financial product includes:

[0023] Obtain the financial product interest lists of multiple sample users;

[0024] Based on the financial product interest list, a first user who is interested in the target financial product, a second user who is interested in the preferred financial product, and a third user who is interested in both the target financial product and the preferred financial product are identified.

[0025] The activity level of each third user is determined based on the financial product interest list, and the similarity contribution factor of each third user is determined based on the activity level, such that the similarity contribution factor is negatively correlated with the activity level.

[0026] A first sum is determined based on the sum of the similarity contribution factors of the third user, a first mean is determined based on the geometric mean of the number of the first user and the number of the second user, and then the product similarity between the target financial product and the preferred financial product is determined based on the quotient of the first sum and the first mean.

[0027] In some embodiments, the step of selecting multiple financial products to be recommended from the target financial products based on the product similarity includes:

[0028] Determine the target user's user preference score for the preferred financial product;

[0029] The interest value of the target user for the target financial product is determined by multiplying the user preference score and the product similarity.

[0030] When the interest value is greater than or equal to a preset first threshold, the corresponding target financial product is determined as the financial product to be recommended, and the recommendation weight of the financial product to be recommended is determined based on the interest value.

[0031] In some embodiments, determining a financial product portfolio based on the optimal cash holding and the financial products to be recommended, and determining the expected return and investment style of the financial product portfolio, includes:

[0032] The target user's financial product asset amount is determined based on the optimal cash holding amount;

[0033] The investment share of each of the financial products to be recommended is determined based on the recommendation weight and the asset amount of the financial products, and then the financial product portfolio is generated based on the financial products to be recommended and the corresponding investment shares.

[0034] The expected return of the financial product portfolio is obtained based on the historical return data of the financial products to be recommended and the investment share forecast.

[0035] The investment style of the financial product portfolio is determined based on the product style of the financial product to be recommended and the investment share.

[0036] In some embodiments, the financial product recommendation method further includes:

[0037] Determine the target similarity between the financial product to be recommended and other financial products in the target financial product category;

[0038] Based on the target similarity, several candidate financial products are selected from the other financial products;

[0039] When the target user is not interested in the financial product to be recommended in the financial product portfolio, the financial product portfolio is adjusted according to the alternative financial products.

[0040] To achieve the above objectives, a second aspect of this application provides a financial product recommendation device, the device comprising:

[0041] The historical transaction analysis module is used to obtain the target user's historical transaction records, predict the target user's optimal cash holding amount based on the historical transaction records, and determine several preferred financial products of the target user based on the historical transaction records.

[0042] The product similarity determination module is used to query multiple preset target financial products and determine the product similarity between each target financial product and the preferred financial product.

[0043] The financial product screening module is used to screen multiple financial products to be recommended from the target financial products based on the product similarity.

[0044] The financial product portfolio determination module is used to determine a financial product portfolio based on the optimal cash holding amount and the financial products to be recommended, and to determine the expected return and investment style of the financial product portfolio, and then generate a financial product recommendation plan based on the financial product portfolio, the expected return and the investment style.

[0045] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for enabling communication between the processor and the memory. When the program is executed by the processor, it implements the financial product recommendation method as described in the first aspect above.

[0046] To achieve the above objectives, a fourth aspect of this application provides a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more programs that can be executed by one or more processors to implement the financial product recommendation method as described in the first aspect above.

[0047] The financial product recommendation method, apparatus, electronic device, and storage medium proposed in this application acquire the target user's historical transaction records, predict the target user's optimal cash holding amount based on the historical transaction records, determine several preferred financial products based on the historical transaction records, then query multiple preset target financial products, determine the product similarity between each target financial product and the preferred financial products, further filter multiple financial products to be recommended based on product similarity, finally determine a financial product portfolio based on the optimal cash holding amount and the financial products to be recommended, determine the expected return and investment style of the financial product portfolio, and then generate a financial product recommendation scheme based on the financial product portfolio, expected return, and investment style. The embodiments of this application predict the optimal cash holding amount and determine preferred financial products based on the user's historical transaction records, filter out financial products to be recommended based on the product similarity between the target financial products and the preferred financial products, and generate a financial product portfolio for recommendation based on the optimal cash holding amount and the financial products to be recommended. This can provide users with financial product recommendation schemes that match their preferences, improve the accuracy of financial product recommendations, avoid blind investment by users, improve the security of user investments, and thus improve user satisfaction and user stickiness. Attached Figure Description

[0048] Figure 1 This is a flowchart of a financial product recommendation method provided in an embodiment of this application;

[0049] Figure 2 yes Figure 1 A flowchart of step S101 in the process;

[0050] Figure 3 yes Figure 1 Another flowchart for step S101 in the process;

[0051] Figure 4 yes Figure 1 A flowchart of step S102 in the process;

[0052] Figure 5 yes Figure 1 A flowchart of step S103 in the process;

[0053] Figure 6 yes Figure 1 A flowchart of step S104 in the process;

[0054] Figure 7 This is another flowchart of the financial product recommendation method provided in the embodiments of this application;

[0055] Figure 8 This is a schematic diagram of the structure of the financial product recommendation device provided in the embodiments of this application;

[0056] Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application;

[0057] Figure 10 This is a schematic diagram of the structure of the storage medium provided in the embodiments of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0059] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0061] First, let's analyze some of the terms used in this application:

[0062] Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0063] Financial products refer to the various vehicles through which funds are transferred, including currencies, gold, foreign exchange, and securities. In other words, these financial products are the objects of buying and selling in the financial market. Supply and demand, through market competition, determine the price of financial products, such as interest rates or yields, ultimately completing the transaction and achieving the purpose of financing. Stocks, futures, options, and insurance policies are examples of financial assets, also called financial instruments or securities.

[0064] User preferences refer to the rational, biased choices users make when considering goods and services. They are a comprehensive result of user cognition, psychological feelings, and rational economic considerations. In daily life, the concept of "preference" is used widely, consciously and unconsciously. People often use their preferences to assist in daily decision-making or simply to express liking or inclinations without actually making a choice. More often, we need to deal with more complex preference analysis situations, such as conditional preferences, where users need to meet certain conditions and prerequisites when making biased choices and intentional judgments.

[0065] As competition intensifies and financial innovation deepens, users are facing increasingly diverse and complex choices of financial products, leading to a rise in the time cost for users to understand various financial products. Without a thorough understanding of different financial products, users instinctively prefer to choose those they are more familiar with for asset appreciation, resulting in high concentration of users on a single type of product and low user stickiness.

[0066] In the industry, research on product portfolio management mainly focuses on the following two aspects:

[0067] 1) Portfolio Strategy: By analyzing overall market trends, formulate a series of effective product portfolio strategies and promote them.

[0068] 2) Product substitution: By analyzing each product in the product portfolio, identify products with problems and declining sales, provide alternative solutions, and optimize the product structure.

[0069] Currently, product portfolio recommendations are mainly made through recommendations from account managers and discussions with users. In the process of product portfolio management, on the one hand, it is impossible to guarantee that all account managers are thoroughly familiar with all products, and they are prone to having fixed ideas when making product portfolios, resulting in unsatisfactory portfolio effects, affecting the accuracy of financial product recommendations, and failing to meet user needs; on the other hand, due to the lack of relevant data support, it is difficult to convince users based solely on experience, which fails to increase user trust and thus affects user stickiness.

[0070] To meet the diverse needs of users and maximize their benefits, it is necessary to provide users with higher-quality and more comprehensive financial services through a scientific and reasonable product portfolio, thereby maximizing competitiveness and profits.

[0071] Based on this, embodiments of this application provide a financial product recommendation method, apparatus, electronic device, and storage medium, aiming to improve the accuracy of financial product recommendations and user stickiness. Embodiments of this application combine a product portfolio recommendation algorithm with a cash-in-stock model. First, based on data such as customer cash inflows and outflows and interest income, the optimal cash holding amount for the user is predicted. Then, financial products to be recommended are screened based on the user's preferred financial products and product similarity. Finally, financial product recommendations are made based on the optimal cash holding amount and the recommended financial products. This ensures the safety of user assets while formulating an optimal financial product portfolio to maximize user returns.

[0072] The financial product recommendation method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the financial product recommendation method in this application embodiment is described.

[0073] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0074] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0075] The financial product recommendation method provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the financial product recommendation method, but is not limited to the above forms.

[0076] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0077] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0078] Figure 1 This is an optional flowchart of the financial product recommendation method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S101 to S104.

[0079] Step S101: Obtain the target user's historical transaction records, predict the target user's optimal cash holding amount based on the historical transaction records, and determine several preferred financial products of the target user based on the historical transaction records.

[0080] Step S102: Query multiple preset target financial products and determine the product similarity between each target financial product and the preferred financial product;

[0081] Step S103: Select multiple financial products to be recommended from the target financial products based on product similarity;

[0082] Step S104: Determine the financial product portfolio based on the optimal cash holding amount and the financial products to be recommended, and determine the expected return and investment style of the financial product portfolio. Then, generate a financial product recommendation plan based on the financial product portfolio, expected return, and investment style.

[0083] Steps S101 to S104, as illustrated in this embodiment, predict the optimal cash holding amount and determine preferred financial products based on the user's historical transaction records. Then, select recommended financial products based on the product similarity between the target financial product and the preferred financial product. Finally, generate a financial product portfolio based on the optimal cash holding amount and the recommended financial products for recommendation. This provides users with a financial product recommendation scheme that matches their preferences, improves the accuracy of financial product recommendations, avoids blind investment by users, enhances the security of user investments, and thus increases user satisfaction and user stickiness.

[0084] Please see Figure 2 In some embodiments, the step of predicting the target user's optimal cash holding amount based on historical transaction records in step S101 may include, but is not limited to, steps S1011 to S1014:

[0085] Step S1011: Determine the target user's current securities information and the first total cash demand in each preset historical period based on historical transaction records;

[0086] Step S1012: Predict the target user's second total cash demand in the target period based on the first total cash demand.

[0087] Step S1013: Determine the target user's transaction costs and opportunity cost rate based on the current securities information;

[0088] Step S1014: Determine the optimal cash holding amount based on the second total cash demand, transaction costs, and opportunity cost rate.

[0089] In step S1011 of some embodiments, the information on the securities currently held by the target user can be determined based on historical transaction records. At the same time, the cash inflow and outflow of the target user can be calculated based on the transaction data of each preset historical period, thereby obtaining the first total cash demand for each preset historical period.

[0090] In step S1012 of some embodiments, the second total cash demand of the target user in the target period (the period in which financial product recommendations are required) can be predicted based on the average value of the first total cash demand. Alternatively, a cash demand prediction model can be pre-trained based on historical data of the target user's total cash demand. The changing trend of the target user's total cash demand can be fitted based on the model, and then the first total cash demand of multiple preset historical periods can be combined to make a prediction to obtain the second total cash demand of the target period.

[0091] In step S1013 of some embodiments, this application embodiment uses a cash inventory model to calculate the optimal cash holding amount for the target user in order to maximize the user's benefits.

[0092] The cash-in-stock model is a method for managing cash. This model analyzes a company's daily cash flow data to calculate the minimum cash reserves required to maintain normal operations. This application applies this model to the practical scenario of recommending financial products, helping companies or individuals develop effective cash management strategies to increase returns while maintaining operational stability. Under the cash-in-stock model, the total cost of holding cash assets includes two aspects: first, the holding cost, also known as opportunity cost, which is the return forgone by holding cash. This cost is typically the interest on marketable securities and is proportional to the cash balance; second, the conversion cost, also known as transaction cost, refers to the fixed costs of cash and marketable securities. This cost is only related to the number of transactions and is independent of the amount of cash held.

[0093] In this embodiment, total cost (T) represents the total cash demand of the enterprise over a certain period; opportunity cost is the investment income lost by the enterprise due to retaining a certain cash balance for daily expenses, i.e., average cash holding * marketable securities interest rate (K); and transaction cost (F) is the interest loss incurred from early withdrawal after the cash has been transferred to capital gains. Based on the relevant theories of the cash-in-stock model, the formula for calculating the optimal cash holding (usable for daily expenses and flexibly accessible during demand) is as follows:

[0094]

[0095] Where C represents the optimal cash holding, T represents the total cost, i.e. the total cash requirement, K represents the interest rate on marketable securities, i.e. the opportunity cost rate, and F represents the transaction cost.

[0096] Based on the current securities information determined by the aforementioned steps, the target user's current holdings of securities, interest rates, and other information can be determined, thereby calculating the securities interest rate, i.e., opportunity cost, as well as transaction costs.

[0097] In step S1014 of some embodiments, the optimal cash holding amount can be calculated by substituting the second total cash demand, transaction costs, and opportunity cost rate obtained in the aforementioned steps into the aforementioned formula.

[0098] It is understood that, based on the target user's historical transaction records and the cash inventory model, the embodiment of this application calculates the target user's optimal cash holding amount, that is, the target user's minimum cash reserve, thereby helping the target user to formulate an effective cash management strategy, improve the accuracy of financial product recommendations, and maximize user benefits.

[0099] Please see Figure 3 In some embodiments, the step S101, which involves determining several preferred financial products of the target user based on historical transaction records, may include, but is not limited to, steps S1015 to S1017:

[0100] Step S1015: Determine the historical purchase frequency, historical purchase volume, and historical holding duration of multiple historical financial products based on historical transaction records;

[0101] Step S1016: Based on the preset weighting coefficients, the historical purchase frequency, historical purchase share and historical holding period are weighted and summed to obtain the user preference score for each historical financial product.

[0102] Step S1017: Select several preferred financial products from historical financial products based on user preference scores.

[0103] In step S1015 of some embodiments, the historical purchase frequency, historical purchase amount and historical holding time of multiple historical financial products purchased by the target user are statistically analyzed based on historical transaction records. It can be understood that the more historical purchase frequency, the larger the historical purchase amount and the longer the historical holding time, the more the target user prefers the historical financial product. Therefore, these statistical data can be used to calculate the user preference score of each historical financial product in the future.

[0104] In step S1016 of some embodiments, weighting coefficients for historical purchase frequency, historical purchase share, and historical holding duration are preset. Based on these weighting coefficients, the historical purchase frequency, historical purchase share, and historical holding duration obtained in the aforementioned steps are weighted and summed to obtain the user preference score for each historical financial product.

[0105] In step S1017 of some embodiments, the top M historical financial products with the highest user preference scores can be selected as preferred financial products, or a preference score threshold can be preset, and historical financial products with user preference scores greater than or equal to the preference score threshold can be selected as preferred financial products.

[0106] It is understood that, according to the embodiments of this application, the historical purchase frequency, historical purchase share and historical holding period of multiple historical financial products are statistically analyzed based on historical transaction records, and user preference scores of each historical financial product are calculated in combination with preset weight coefficients, so as to screen out the target user's preferred financial products for subsequent screening of financial products to be recommended.

[0107] Please see Figure 4 In some embodiments, the step of determining the product similarity between each target financial product and the preferred financial product in step S102 may include, but is not limited to, steps S1021 to S1024:

[0108] Step S1021: Obtain the financial product interest list of multiple sample users;

[0109] Step S1022: Based on the financial product interest list, determine the first user who is interested in the target financial product, the second user who is interested in the preferred financial product, and the third user who is interested in both the target financial product and the preferred financial product;

[0110] Step S1023: Determine the activity level of each third user based on the financial product interest list, and determine the similarity contribution factor of each third user based on the activity level, so that the similarity contribution factor is negatively correlated with the activity level.

[0111] Step S1024: Determine the first sum based on the sum of the similarity contribution factors of the third user, determine the first mean based on the geometric mean of the number of the first user and the number of the second user, and then determine the product similarity between the target financial product and the preferred financial product based on the quotient of the first sum and the first mean.

[0112] In step S1021 of some embodiments, a list of financial product interests of multiple sample users collected in advance is obtained, which contains financial products that each sample user is interested in.

[0113] In step S1022 of some embodiments, based on the financial product interest list, it can be determined whether each sample user is interested in the target financial product and whether they are interested in the preferred financial product. This allows for the selection of a first user interested in the target financial product (regardless of interest in the preferred financial product), a second user interested in the preferred financial product (regardless of interest in the target financial product), and a third user interested in both the target and preferred financial products. It should be noted that there may be overlap between the first, second, and third users in this embodiment; they are not entirely independent user groups.

[0114] In this embodiment, an item-based collaborative filtering algorithm (ItemCF) is used to calculate the similarity between the target financial product and the preferred financial product. The ItemCF algorithm does not directly calculate similarity based on the attributes of the items themselves, but rather by analyzing user behavior. Taking financial products as an example, when many users are simultaneously interested in both financial product A (such as a basic deposit product) and financial product B (such as a value-added deposit product), it indicates that financial product A and financial product B have a certain degree of similarity.

[0115] The above analysis shows that the similarity between two financial products can be calculated using the following formula:

[0116]

[0117] Where N(μ) represents the number of users who like financial product μ, |N(μ)| represents the number of users who like financial product μ, N(μ)∩N(v) represents the number of users who like both financial products μ and v, and ||N(μ)∩N(v)| represents the number of users who like both financial products μ and v. Therefore, the similarity W between financial product v and financial product μ in the above formula is... μv This can be understood as: how many users who like financial product μ also like financial product v? If all users who like financial product μ also like financial product v, then W... μv =1.

[0118] However, if the financial product v is a popular financial product that many users like, then the numerator and denominator in the above formula will be very close, and W will be... μv This would result in a penalty value very close to 1, meaning any financial product would have a high similarity to popular financial products. This would cause the ItemCF algorithm to consistently recommend popular financial products, impacting the comprehensiveness and diversity of recommendations. Therefore, it's necessary to increase the penalty value for popular financial products to mitigate the likelihood that they will be similar to many other financial products. The following formula can be used:

[0119]

[0120] The improvement in the above formula is that the denominator is replaced by the geometric mean of |N(μ)| and |N(v)|. If |N(v)| is large, the denominator will also become large, which is equivalent to penalizing the weight of popular financial product v and reducing the possibility that popular financial product v will be similar to many other items.

[0121] In the above formula, each user's interest list contributes equally to the similarity of financial products. However, in practical applications, an active user's interest list may contain most financial products, but these products may not be related. Obviously, the interest list of an active user will affect the similarity assessment of financial products. Therefore, the contribution of active users to the similarity of financial products should be less than that of inactive users. This application's embodiment uses a similarity contribution factor to modify the similarity formula, as follows:

[0122]

[0123] Where i∈N(μ)∩N(v) represents user i who likes both financial products μ and v, N(i) represents the financial products that user i likes, and |N(i)| represents the number of financial products that user i likes, which is also the activity level of user i. The similarity contribution factor is represented by the base of the logarithmic function log(1+|N(i)|), which can be e or 10, and is not limited here.

[0124] The above formula sums the similarity contribution factors of user i. By replacing ‖N(μ)∩N(v)‖ in the previous formula, it can be seen that the higher the user's activity level, the smaller the similarity contribution factor, which can reduce the contribution of active users to the similarity of financial products.

[0125] In step S1023 of some embodiments, the activity level of the third user can be determined based on the number of financial products that the third user is interested in in the financial product interest list. The more financial products the third user is interested in, the higher the activity level of the third user.

[0126] In step S1024 of some embodiments, a first sum is determined based on the sum of the similarity contribution factors of the third user, which is the numerator of the modified similarity formula. A first mean is determined based on the geometric mean of the number of the first user and the number of the second user, which is the denominator of the modified similarity formula. The quotient of the two is used as the product similarity between the target financial product and the preferred financial product.

[0127] It is understood that the embodiments of this application use an item-based collaborative filtering algorithm (ItemCF) to calculate the similarity between the target financial product and the preferred financial product. This fully considers the preferences of different users when selecting financial products, improves the accuracy of financial product similarity calculation, and thus improves the accuracy of financial product recommendations.

[0128] Please see Figure 5 In some embodiments, step S103 may include, but is not limited to, steps S1031 to S1033:

[0129] Step S1031: Determine the target user's user preference score for preferred financial products;

[0130] Step S1032: Determine the target user's interest value in the target financial product based on the product of user preference score and product similarity.

[0131] Step S1033: When the interest value is greater than or equal to the preset first threshold, the corresponding target financial product is determined as the financial product to be recommended, and the recommendation weight of the financial product to be recommended is determined according to the interest value.

[0132] In step S1031 of some embodiments, referring to the descriptions in steps S1015 and S1016, the historical purchase frequency, historical purchase share, and historical holding duration of multiple historical financial products purchased by the target user are statistically analyzed based on historical transaction records. Weighting coefficients for the historical purchase frequency, historical purchase share, and historical holding duration are preset. The obtained historical purchase frequency, historical purchase share, and historical holding duration are weighted and summed according to the weighting coefficients to obtain the user preference score for each historical financial product.

[0133] In step S1032 of some embodiments, the product similarity between the target financial product v and the preferred financial product μ can be expressed as W. μv The user preference score of target user i for the preferred financial product μ can be expressed as R. μi Then W μv R μi P can represent the interest value of target user i for target financial product v. vi .

[0134] In this embodiment of the application, considering that target user i may have multiple preferred financial products μ, and that the same target financial product v may have multiple preferred financial products μ with high similarity, W can be calculated for both the target financial product v and each preferred financial product μ. μv R μi Then, sum them up to obtain the interest value of target user i for target financial product v. The specific formula is as follows:

[0135]

[0136] Where S(v,K) represents K financial products that are highly similar to the target financial product v, N(i) represents the preferred financial product of the target user i, and μ∈S(v,K)∩N(i) represents the preferred financial product that is highly similar to the target financial product v.

[0137] In step S1032 of some embodiments, after calculating the interest value of the target user for each target financial product, the interest value is compared with a preset first threshold, and the target financial product with an interest value greater than or equal to the first threshold is selected as the financial product to be recommended. At the same time, the recommendation weight of the financial product to be recommended can be set based on the size of the interest value, so that the financial product to be recommended with a larger interest value can obtain a larger recommendation weight.

[0138] It can be recognized that the embodiments of this application calculate the target user's interest value in the target financial product based on the product similarity between the target financial product and the preferred financial product, as well as the user preference score of the preferred financial product. This can accurately determine the target user's degree of interest in the target financial product, thereby accurately screening out the financial products that the target user is interested in and improving the accuracy of financial product recommendations.

[0139] Please see Figure 6 In some embodiments, the step S104, which involves determining a financial product portfolio based on the optimal cash holding and the recommended financial products, and determining the expected return and investment style of the financial product portfolio, may include, but is not limited to, steps S1041 to S1044:

[0140] Step S1041: Determine the target user's financial product asset limit based on the optimal cash holding amount;

[0141] Step S1042: Determine the investment share of each financial product to be recommended based on the recommendation weight and the asset amount of the financial product, and then generate a financial product portfolio based on the financial products to be recommended and the corresponding investment shares.

[0142] Step S1043: Based on the historical return data and investment share forecast of the financial products to be recommended, the expected return of the financial product portfolio is obtained;

[0143] Step S1044: Determine the investment style of the financial product portfolio based on the product style and investment share of the financial products to be recommended.

[0144] In step S1041 of some embodiments, the asset amount of the financial product can be determined based on the optimal cash holding amount and the total assets of the target user.

[0145] In step S1042 of some embodiments, the investment ratio of each financial product to be recommended can be determined according to the recommendation weight of the aforementioned steps, and the investment share of each financial product to be recommended can be determined by combining the asset amount of the financial product, thereby obtaining a financial product portfolio that includes the financial products to be recommended and their corresponding investment shares.

[0146] In step S1043 of some embodiments, after recommending a financial product portfolio to the user's assets based on the optimal cash holding amount, the expected increase in interest income of the financial product portfolio in the next three months can be predicted based on the historical return data and investment share of the financial products to be recommended, i.e., the expected return of the financial product portfolio.

[0147] In step S1044 of some embodiments, the corresponding risk coefficient can be determined according to the product style of each financial product to be recommended. For example, the risk coefficients of conservative, stable, balanced, aggressive and volatile financial products increase in sequence. Then, the risk coefficients of each financial product to be recommended and the investment share ratio are weighted and summed to obtain the average risk coefficient of the financial product portfolio. The investment style of the financial product portfolio can be determined based on the average risk coefficient, which can also be divided into conservative, stable, balanced, aggressive and volatile types.

[0148] It can be recognized that, according to the recommendation weight of the financial products to be recommended, the embodiments of this application plan and allocate the target user's financial product asset amount to obtain a financial product portfolio, and determine its expected returns and investment style, thereby obtaining a financial product recommendation scheme that meets the user's needs, improving user satisfaction and user stickiness.

[0149] Please see Figure 7 In some embodiments, the financial product recommendation method provided in this application may also include, but is not limited to, steps S105 to S107:

[0150] Step S105: Determine the target similarity between the financial product to be recommended and other financial products in the target financial product category;

[0151] Step S106: Select several candidate financial products from other financial products based on target similarity;

[0152] Step S107: When the target user is not interested in the financial products to be recommended in the financial product portfolio, the financial product portfolio is adjusted according to the alternative financial products.

[0153] In step S105 of some embodiments, after selecting the financial products to be recommended, the remaining financial products in the target financial products are regarded as other financial products, and the target similarity between the financial products to be recommended and other financial products is calculated using the aforementioned similarity calculation formula.

[0154] In step S106 of some embodiments, several candidate financial products are selected from other financial products based on the target similarity. Specifically, other financial products with the highest target similarity to each financial product to be recommended can be selected as candidate financial products corresponding to that financial product to be recommended. Alternatively, multiple other financial products with the highest sum of target similarities to all financial products to be recommended can be selected as candidate financial products for the overall financial product portfolio.

[0155] In step S107 of some embodiments, if the target user is not satisfied with a certain financial product to be recommended in the recommended financial product portfolio, he / she can select the alternative financial product corresponding to the financial product to be recommended to replace it, and recalculate the investment share of each financial product in the financial product portfolio. Alternatively, he / she can select the alternative financial products of the entire financial product portfolio to replace them in turn, and recalculate the investment share of each financial product in the financial product portfolio.

[0156] In this embodiment of the application, by selecting alternative financial products, the financial product portfolio can be adjusted when the user is not satisfied with the financial product portfolio, thereby further improving user satisfaction and user stickiness.

[0157] In some optional embodiments, for newly opened accounts, in order to increase user stickiness in the later stage, the design of financial product portfolios must first be user-demand-oriented, including understanding the user's investment objectives, investment period, investment amount, investment preferences, etc., and then designing financial product portfolios that meet the user's needs through financial product recommendation algorithms, and tracking the returns in real time; after the user has had an account for a period of time, the financial product recommendation method of this application embodiment can also be used to adjust and recommend financial product portfolios to the user based on the user's historical transaction records.

[0158] It can be recognized that the embodiments of this application predict the optimal cash holding amount and determine the preferred financial products based on the user's historical transaction records, screen the financial products to be recommended based on the product similarity between the target financial products and the preferred financial products, and generate a financial product portfolio for recommendation based on the optimal cash holding amount and the financial products to be recommended. This can provide users with a financial product recommendation scheme that matches their preferences, improve the accuracy of financial product recommendations, avoid blind investment by users, improve the security of user investment, and thus improve user satisfaction and user stickiness.

[0159] The embodiments of this application can help users improve the flexibility and security of financial products. By predicting and analyzing the cash inflows and outflows of financial products, users can better understand the risks and returns of financial products, thereby providing users with more diversified financial products to meet different customer needs. In addition, the embodiments of this application can help users increase the returns of financial products, thereby maximizing user benefits and improving user satisfaction. They can also help financial institutions improve the transparency and fairness of financial products, enhance users' trust in financial institutions, and increase user stickiness.

[0160] Please see Figure 8 This application also provides a financial product recommendation device, which can realize the above-mentioned financial product recommendation. The financial product recommendation device includes:

[0161] The historical transaction analysis module is used to obtain the target user's historical transaction records, predict the target user's optimal cash holding amount based on the historical transaction records, and determine several preferred financial products of the target user based on the historical transaction records.

[0162] The product similarity determination module is used to query multiple preset target financial products and determine the product similarity between each target financial product and the preferred financial product.

[0163] The financial product screening module is used to filter multiple financial products to be recommended from the target financial products based on product similarity.

[0164] The financial product portfolio determination module is used to determine a financial product portfolio based on the optimal cash holding and the financial products to be recommended, and to determine the expected return and investment style of the financial product portfolio. Then, it generates a financial product recommendation plan based on the financial product portfolio, expected return, and investment style.

[0165] The specific implementation of this financial product recommendation device is basically the same as the specific implementation of the above-described financial product recommendation method, and will not be repeated here.

[0166] This application also provides an electronic device, which includes: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for communication between the processor and the memory. When the program is executed by the processor, it implements the aforementioned financial product recommendation method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0167] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0168] The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0169] The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and is called and executed by the processor 901 using the financial product recommendation method of the embodiments of this application.

[0170] The input / output interface 903 is used to implement information input and output;

[0171] The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0172] Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904);

[0173] The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0174] Please see Figure 10 This application also provides a storage medium, which is a computer-readable storage medium for computer-readable storage. The storage medium stores one or more programs 1001, which can be executed by one or more processors to implement the above-mentioned financial product recommendation method.

[0175] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0176] The financial product recommendation method, apparatus, electronic device, and storage medium provided in this application acquire the historical transaction records of a target user, predict the target user's optimal cash holding amount based on the historical transaction records, determine several preferred financial products based on the historical transaction records, then query multiple preset target financial products, determine the product similarity between each target financial product and the preferred financial products, further filter multiple financial products to be recommended based on the product similarity, finally determine a financial product portfolio based on the optimal cash holding amount and the financial products to be recommended, determine the expected return and investment style of the financial product portfolio, and then generate a financial product recommendation scheme based on the financial product portfolio, expected return, and investment style. This application embodiment predicts the optimal cash holding amount and determines preferred financial products based on the user's historical transaction records, filters out financial products to be recommended based on the product similarity between the target financial products and the preferred financial products, and generates a financial product portfolio for recommendation based on the optimal cash holding amount and the financial products to be recommended. This provides users with financial product recommendation schemes that match their preferences, improves the accuracy of financial product recommendations, avoids blind investment by users, improves the security of user investments, and thus improves user satisfaction and user stickiness.

[0177] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0178] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0179] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0180] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0181] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0182] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0183] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0184] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0185] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0186] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0187] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A financial product recommendation method, characterized in that: The method comprises: Acquire historical transaction records of a target user, predict the optimal cash holding amount of the target user based on the historical transaction records, and determine a number of preferred financial products of the target user based on the historical transaction records; Querying a plurality of preset target financial products to determine the product similarity between each of the target financial products and the preferred financial product; Filtering the target financial products according to the product similarities to obtain a plurality of financial products to be recommended; A financial product portfolio is determined based on the optimal cash holding amount and the financial products to be recommended, and the expected return and investment style of the financial product portfolio are determined. A financial product recommendation plan is then generated based on the financial product portfolio, the expected return and the investment style, and the financial product recommendation plan is recommended to the target user.

2. The financial product recommendation method according to claim 1, characterized in that: The predicting the optimal cash holding amount of the target user according to the historical transaction records includes: Determine the current securities information of the target user and the first total cash demand in each preset historical period according to the historical transaction records; Predicting a second total cash demand of the target user in a target period according to the first total cash demand; Determining the transaction cost and opportunity cost rate of the target user according to the current securities information; The optimal cash holding amount is determined according to the second total cash demand, the transaction cost and the opportunity cost rate.

3. The financial product recommendation method according to claim 1, characterized in that: The determining of the plurality of preferred financial products of the target user according to the historical transaction records includes: Determine the historical purchase times, historical purchase shares, and historical holding durations of multiple historical financial products based on the historical transaction records; The historical purchase times, historical purchase shares and historical holding durations are weighted and summed according to a preset weight coefficient to obtain a user preference score for each of the historical financial products; A plurality of the preferred financial products are screened out from the historical financial products according to the user preference scores.

4. The financial product recommendation method according to claim 1, characterized in that: The determining of the product similarity between each of the target financial products and the preferred financial product includes: Obtain a list of financial product interests of multiple sample users; Determine, according to the financial product interest list, a first user who is interested in the target financial product, a second user who is interested in the preferred financial product, and a third user who is interested in both the target financial product and the preferred financial product; determining the activity of each of the third users according to the financial product interest list, and determining a similarity contribution factor of each of the third users according to the activity, such that the similarity contribution factor is negatively correlated with the activity; A first sum is determined according to the sum of the similarity contribution factors of the third users, a first mean is determined according to the geometric mean of the number of the first users and the number of the second users, and then the product similarity between the target financial product and the preferred financial product is determined according to the quotient of the first sum and the first mean.

5. The financial product recommendation method according to claim 1, characterized in that: The step of screening the target financial products according to the product similarity to obtain a plurality of financial products to be recommended includes: Determining a user preference score of the target user for the preferred financial product; Determining the interest value of the target user in the target financial product according to the product of the user preference score and the product similarity; When the interest value is greater than or equal to a preset first threshold, the corresponding target financial product is determined to be the financial product to be recommended, and a recommendation weight of the financial product to be recommended is determined according to the interest value.

6. The financial product recommendation method according to claim 5, characterized in that: The step of determining a financial product portfolio based on the optimal cash holding amount and the financial products to be recommended, and determining the expected return and investment style of the financial product portfolio includes: Determining the target user's financial product asset amount based on the optimal cash holding amount; Determine the investment share of each of the financial products to be recommended according to the recommendation weight and the asset amount of the financial products, and then generate the financial product portfolio according to the financial products to be recommended and the corresponding investment shares; Obtaining the expected return of the financial product portfolio according to the historical return data of the financial product to be recommended and the investment share forecast; The investment style of the financial product portfolio is determined according to the product style of the to-be-recommended financial product and the investment share.

7. The financial product recommendation method according to any one of claims 1 to 6, characterized in that: The financial product recommendation method further includes: Determining target similarity between the to-be-recommended financial product and other financial products in the target financial product; Filtering a number of candidate financial products from the other financial products according to the target similarity; When the target user is not interested in the to-be-recommended financial product in the financial product portfolio, the financial product portfolio is adjusted according to the candidate financial product.

8. A financial product recommendation device, characterized in that: The device comprises: A historical transaction analysis module, used to obtain historical transaction records of a target user, predict the optimal cash holding amount of the target user based on the historical transaction records, and determine a number of preferred financial products of the target user based on the historical transaction records; A product similarity determination module, used to query a plurality of preset target financial products and determine the product similarity between each of the target financial products and the preferred financial product; A financial product screening module, used for screening a plurality of financial products to be recommended from the target financial products according to the product similarity; A financial product combination determination module is used to determine a financial product combination based on the optimal cash holding amount and the financial products to be recommended, and to determine the expected return and investment style of the financial product combination, and then generate a financial product recommendation plan based on the financial product combination, the expected return and the investment style.

9. An electronic device, characterized in that: The electronic device includes a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory. When the program is executed by the processor, the steps of the financial product recommendation method as described in any one of claims 1 to 7 are realized.

10. A storage medium, the storage medium being a computer-readable storage medium, used for computer-readable storage, characterized in that: The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the financial product recommendation method as described in any one of claims 1 to 7.

Images