Product recommendation method and device, electronic equipment and computer program product

Abstract

The invention discloses a product recommendation method and device, electronic equipment and a computer program product. The method comprises the following steps: acquiring financial management information of a user in a preset period; inputting the financial management information into a transaction model to obtain an investment portrait of the user, the transaction model being obtained by training a plurality of groups of training samples, each group of training samples comprising historical financial management information and a historical investment portrait, and the investment portrait being used for representing an investment habit of the user; a predicted transaction behavior of the user is determined based on the investment portrait, and the predicted transaction behavior comprises at least one of deposit, mortgage, stock transaction and fund transaction; and determining a to-be-recommended product of the target financial institution based on the predicted transaction behavior, and pushing the to-be-recommended product to a client of the user. Through the method and the device, the problem that the recommended product is not matched with the product purchase tendency of the user in the related technology is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and more specifically, to a product recommendation method, device, electronic device, and computer program product. Background Art

[0002] There are many types of financial products and differentiated financial needs. Financial institutions need to provide targeted and personalized financial product recommendations based on customer characteristics and preferences in order to improve the success rate of financial product recommendations and promote the development of financial business.

[0003] In related technologies, financial product recommendations are made by analyzing customers' historical transaction behaviors based on information such as customer historical transaction content, transaction amount, and transaction frequency, predicting customers' financial product purchase intentions, and making product recommendations. However, this product recommendation method has certain limitations. For customers, the recommended financial products do not match the products they need, resulting in a low success rate for financial product recommendations.

[0004] Currently, no effective solution has been proposed to address the problem that the products recommended in related technologies do not match the user's product purchasing preferences. Summary of the invention

[0005] The main purpose of the present application is to provide a product recommendation method, device, electronic device and computer program product to solve the problem that the recommended products in the related art do not match the user's product purchasing tendency.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a product recommendation method is provided. The method comprises: obtaining the financial information of the user within a preset period, wherein the financial information comprises at least one of the following: identity information, transaction information and investment information; inputting the financial information into the transaction model to obtain the investment portrait of the user, wherein the transaction model is trained by multiple groups of training samples, each group of training samples comprises historical financial information and historical investment portraits, and the investment portrait is used to characterize the investment habits of the user; determining the predicted transaction behavior of the user based on the investment portrait, wherein the predicted transaction behavior comprises at least one of the following: deposit, mortgage, stock transaction and fund transaction; determining the products to be recommended of the target financial institution based on the predicted transaction behavior, and pushing the products to be recommended to the user's client.

[0007] Optionally, determining the user's predicted transaction behavior based on the investment portrait includes: extracting the user's investment cycle, investment behavior, investment products and investment results from the investment portrait; determining a predicted transaction date for the predicted transaction behavior based on the investment cycle, determining the user's preferred transaction type based on the investment behavior, and determining a predicted transaction amount based on the investment products and investment results; and determining the predicted transaction behavior on the predicted transaction date through the transaction type and predicted transaction amount.

[0008] Optionally, determining the products to be recommended of the target financial institution based on the predicted transaction behavior includes: screening financial products that match both the transaction type and the predicted transaction amount from the target financial institution to obtain a group of financial products; obtaining historical financial products purchased by the user, and determining the similarity between each financial product in the group of financial products and the historical financial products; removing financial products whose similarity is less than a first similarity threshold from the group of financial products to obtain an updated group of financial products; and determining the updated group of financial products as the products to be recommended.

[0009] Optionally, after pushing the product to be recommended to the user's client, the method also includes: determining a recommendation result for the product to be recommended, wherein the recommendation result includes whether the product has been purchased or not; if the recommendation result is that the product has been purchased, determining the target financial product purchased by the user, and updating the user's financial information based on the purchase record of the target financial product.

[0010] Optionally, after determining the recommendation result of the product to be recommended, the method also includes: if the recommendation result is not purchased, detecting whether the user purchases other financial products, wherein other financial products are financial products other than the product to be recommended; if the user is not detected to have purchased other financial products, updating the user's financial information at preset time intervals to obtain updated financial information; if the similarity between the updated financial information and the financial information within a preset period is less than or equal to a second similarity threshold, determining an updated product to be recommended based on the updated financial information, and pushing the updated product to be recommended to the user's client.

[0011] Optionally, after determining the recommendation result of the product to be recommended, the method also includes: if the recommendation result is not purchased, detecting whether the user purchases other financial products, wherein other financial products are financial products other than the product to be recommended; if it is detected that the user purchases other financial products, determining whether the other financial products are products of the target financial institution; if the other financial products are products of the target financial institution, updating the user's financial information based on the purchase records of the other financial products; if the other financial products are not products of the target financial institution, obtaining the financial products in the target financial institution with the greatest similarity to other financial products to obtain similar financial products; and updating the user's financial information based on the product information of the similar financial products.

[0012] Optionally, the transaction model is trained in the following manner: obtaining the transaction records of the target financial institution, extracting each user's historical financial management information and historical investment portrait from the transaction records; determining each user's historical financial management information and historical investment portrait as a group of training samples, and obtaining multiple groups of training samples; training a machine learning model based on the multiple groups of training samples to obtain a transaction model.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a product recommendation device is provided. The device includes: an acquisition unit, which is used to acquire the financial information of the user within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information and investment information; an input unit, which is used to input the financial information into the transaction model to obtain the investment portrait of the user, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the investment habits of the user; a first determination unit, which is used to determine the predicted transaction behavior of the user based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposits, mortgages, stock transactions and fund transactions; a second determination unit, which is used to determine the target financial institution's products to be recommended based on the predicted transaction behavior, and push the recommended products to the user's client.

[0014] In an embodiment of the present application, a method is adopted in which the financial information of a user within a preset period is obtained, wherein the financial information includes at least one of the following: identity information, transaction information and investment information; the financial information is input into a transaction model to obtain an investment portrait of the user, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the investment habits of the user; the predicted transaction behavior of the user is determined based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposits, mortgages, stock transactions and fund transactions; based on the predicted transaction behavior, a target financial institution's to-be-recommended products are determined, and the to-be-recommended products are pushed to the user's client, the user's predicted transaction behavior is determined by analyzing the user's financial information, and then the target financial institution's to-be-recommended products are determined based on the predicted transaction behavior, and the products are recommended to the user, thereby achieving the purpose of personalized recommendation of financial products based on the user's predicted transaction behavior, thereby achieving the technical effect of improving the matching degree between the recommended products and the user's product purchase tendency, and then solving the technical problem of mismatch between the recommended products and the user's product purchase tendency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0016] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a product recommendation method is shown;

[0017] Figure 2 is a flowchart of a product recommendation method provided according to an embodiment of the present application;

[0018] Figure 3is a schematic diagram of a product recommendation device provided according to an embodiment of the present application;

[0019] Figure 4 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] It should be noted that the collected information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with relevant laws, regulations and standards, necessary confidentiality measures are taken, and public order and good customs are not violated, and corresponding operation entrances are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or institutions to provide users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making; if the user chooses to refuse, the expert decision-making process will be entered.

[0023] Example 1

[0024] According to an embodiment of the present application, a method embodiment of a product recommendation is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] The method embodiment provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing a product recommendation method. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more (shown as 102a, 102b, ..., 102n in the figure) processors 102 (the processor 102 may include but is not limited to a processing device such as an MCU (Microcontroller Unit) or an FPGA (Field-Programmable Gate Array), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a USB (Universal Serial Bus) port (which may be included as one of the ports of a BUS (Business)), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0026] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0027] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the product recommendation method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned product recommendation method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0029] The display may be, for example, a touch screen liquid crystal display that enables a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0030] In the above operating environment, the present application provides a product recommendation method. Figure 2 is a flowchart of a product recommendation method provided according to an embodiment of the present application, such as Figure 2 As shown, the method includes:

[0031] Step S201, obtaining the user's financial information within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information and investment information.

[0032] In step S201, the preset period can be set to one day, one week or one month, etc., and the identity information can include the user's name, age, gender, occupation and educational background, etc. This part of information can be read from the customer data database of the financial institution, or obtained from a third-party data platform with the user's consent.

[0033] Transaction information refers to all financial transaction records of users within a preset period, which may include but are not limited to deposits, withdrawals, transfers, consumption, loans, repayments, etc. This information can be collected in real time or periodically through the transaction system of financial institutions. For example, the transaction system of financial institutions can automatically record the time, amount, transaction type and counterparty of each transaction. Investment information may include stock trading, bond holdings, fund investment, insurance purchases, gold trading, etc. This information can be provided by the investment platform of financial institutions, or obtained from the user's securities account, fund account, etc.

[0034] Step S202, input the financial information into the transaction model to obtain the user's investment portrait, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the user's investment habits.

[0035] In step S202, the transaction model can be an algorithmic framework built based on machine learning or deep learning technology to analyze and predict user investment habits. The input of the transaction model is the user's financial information, and the output is the investment portrait. The transaction model can be designed as a supervised learning model, where the goal is to predict the user's investment behavior or preference, or it can be an unsupervised learning model to discover potential patterns in the user's investment behavior. Training the transaction model requires a large amount of historical financial information and historical investment portraits as training data. Use machine learning algorithms (such as logistic regression, support vector machine, random forest, neural network, etc.) to train the extracted features and corresponding historical investment portraits, and optimize the model parameters to improve the prediction accuracy of the transaction model.

[0036] After the trading model training is completed, the user's real-time financial information can be input into the trading model to obtain the user's investment profile. The investment profile reflects the user's investment preferences and can also include more detailed information such as investment risk level, investment goals, and investment time range. Through investment profiles, financial institutions can more accurately understand customers' investment needs and behavior patterns, thereby providing more personalized and accurate investment advice and product recommendations.

[0037] For example, if a user's trading model output shows that he or she tends to invest regularly, prefers low-risk products, has a long investment cycle, etc., then his or her investment profile can be described as a "conservative long-term investor". Financial institutions can recommend investment portfolios that match this profile, such as bond funds, money market funds, or time deposit products.

[0038] Step S203, determining the user's predicted transaction behavior based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposit, mortgage, stock transaction and fund transaction.

[0039] In step S203, the user's possible transaction behavior at a future time point is determined based on the investment profile (i.e., predicted transaction behavior). The investment profile may include key information such as the user's investment preference, risk tolerance, investment goal, and investment cycle, which is obtained through transaction model analysis based on the user's historical financial information and behavior patterns. For example, a user's investment profile may characterize him as a "conservative investor" who prefers low-risk products and tends to invest in the long term.

[0040] Identify features related to various types of trading behaviors based on the user's current investment profile. For example, for a "conservative investor", the user may prefer deposits and fund transactions, while for a "risk-taking investor", the user may prefer stock transactions and mortgages. Analyze the user's past trading behavior patterns in similar periods to find patterns that match the current investment profile. For example, if historical data shows that at the end of each quarter, the conservative investor makes a fixed deposit or purchases a fixed-income fund, then it can be predicted that the user may repeat these behaviors at the end of the upcoming quarter.

[0041] In addition to the user's personal investment profile, it is also necessary to consider the impact of factors such as market conditions and economic environment on the user's trading behavior. For example, when the market is highly volatile, even if the user's historical preference points to stock trading, it may turn to fund trading or deposits due to reduced risk appetite. Combined with the user's investment profile, historical trading patterns and current market environment, use statistical learning, machine learning and other methods to predict the user's future trading behavior. For example, based on the user's preference for low-risk products and investment habits at the end of the quarter, predict the possibility of making a fixed deposit or purchasing a fixed-income fund at the end of the next quarter. Calculate the probability for each possible predicted trading behavior to help financial institutions understand the trading behavior that users may take and its likelihood. For example, predict that users have an 80% chance of conducting fund transactions and a 20% chance of conducting stock transactions.

[0042] For example, for a user with a stable income and low risk appetite, whose investment profile is displayed as a "stable long-term investor", the financial institution's prediction model can analyze his income cycle, consumption habits and investment goals, and predict that he is more likely to make regular deposits or purchase low-risk funds during peak income periods (such as payday or bonus payment days). If a user's investment profile shows that he is a "market analysis investor" who often pays attention to stock market dynamics, the prediction model can combine recent market trends, the user's historical stock trading records and his social behavior (such as activeness in financial forums) to predict that he is likely to trade stocks when specific opportunities arise in the market.

[0043] Step S204, determining the target financial institution's products to be recommended based on the predicted transaction behavior, and pushing the products to be recommended to the user's client.

[0044] In step S204, first, the financial institution needs to establish a product library, which contains all the recommended financial products and their characteristics. Product characteristics may include but are not limited to product type (such as deposits, stocks, funds, insurance, etc.), risk level, expected return, investment cycle, minimum investment threshold, etc. Then, according to the characteristics of the product and the user's predicted transaction behavior, establish product matching rules. For example, if the predicted transaction behavior is a deposit, the matching rules may give priority to fixed deposits or high-interest savings accounts; if it is a stock transaction, stock funds or direct investment in stocks may be recommended. In addition, it is also possible to further match the recommended products that meet the predicted transaction date and predicted transaction amount of the predicted transaction behavior. Using the above product matching rules, the user's predicted transaction behavior is compared with the product characteristics in the product library to filter out a list of products that meet the predicted transaction behavior.

[0045] Furthermore, the product list can be adjusted and sorted based on more specific information such as the user's personal preferences, risk tolerance, financial status, etc. to determine the most suitable recommended products for the user. Personalized product recommendation information is generated based on the user's investment profile and predicted trading behavior. Product recommendation information can include a detailed description of the product, expected returns, risk warnings, operating instructions, etc., to improve the user's understanding and trust in the product.

[0046] The recommended product information is pushed to the user through the user's client (such as mobile application, online banking, email or SMS, etc.). The push method can take into account the user's habit of receiving information, such as non-working hours on weekdays, weekends, etc.

[0047] For example, the predicted transaction behavior is a fixed deposit. Based on the user's investment profile (conservative investor), the goal is to recommend low-risk, fixed-income products. Therefore, financial institutions will select products to be recommended that meet these conditions from the product library, such as a 1-year fixed deposit or a low-risk bond fund.

[0048] Predicted trading behavior is stock trading. Analyze the user's social behavior and market trends to determine that they have a high degree of attention to the stock market and have a certain risk tolerance. Financial institutions can recommend stock funds or platforms that provide stock trading services, and push relevant information, including market analysis, investment advice and operation guides, via email or application messages when favorable opportunities appear in the market.

[0049] The product recommendation method provided in the embodiment of the present application obtains the financial information of the user within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information and investment information; inputs the financial information into the transaction model to obtain the investment portrait of the user, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the investment habits of the user; determines the user's predicted transaction behavior based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposits, mortgages, stock transactions and fund transactions; determines the target financial institution's to-be-recommended products based on the predicted transaction behavior, and pushes the to-be-recommended products to the user's client, determines the user's predicted transaction behavior by analyzing the user's financial information, and then determines the target financial institution's to-be-recommended products based on the predicted transaction behavior, and recommends the products to the user, thereby achieving the purpose of personalized recommendation of financial products based on the user's predicted transaction behavior, thereby achieving the technical effect of improving the matching degree between the recommended products and the user's product purchase tendency, and then solving the technical problem of mismatch between the recommended products and the user's product purchase tendency.

[0050] After determining the investment portrait, the user's predicted transaction behavior is determined based on the investment portrait. Optionally, in the product recommendation method provided in the embodiment of the present application, determining the user's predicted transaction behavior based on the investment portrait includes: extracting the user's investment cycle, investment behavior, investment products and investment results from the investment portrait; determining the predicted transaction date of the predicted transaction behavior based on the investment cycle, determining the user's preferred transaction type based on the investment behavior, and determining the predicted transaction amount based on the investment product and investment results; determining the predicted transaction behavior on the predicted transaction date through the transaction type and the predicted transaction amount.

[0051] In some examples, the periodic characteristics (investment cycle) of users making specific types of investments are extracted from the investment portrait, such as the user's habit of buying funds at the end of each quarter, or making fixed investments on a fixed date every month. Identify the user's past investment behavior patterns, such as a preference for fixed deposits, stock trading, and fund investments. Record the types of products and preferences that users have invested in in the past, such as stocks, bonds, and hybrid funds. Analyze the user's investment returns (i.e., investment results), including yields, risk assessments, etc., to understand the user's investment successes and failures.

[0052] Based on the user's investment cycle, the date on which the user may make the next transaction can be predicted. For example, if the user's fund fixed investment cycle is the 10th of each month, then the predicted transaction date is the 10th of the next month. Based on the user's historical investment behavior in the investment portrait, the user's preferred transaction type can be determined. For example, if the user has frequent stock trading records, the predicted transaction type may be biased towards stock trading. By analyzing the user's past investment amount, income, and capital inflow and outflow in investment products, the amount that the user may invest in the next transaction can be predicted. For example, if the user has invested an average of 5,000 yuan each time in the past fund fixed investment, and the account has a large inflow of funds recently, the predicted transaction amount may be higher than the average.

[0053] Comprehensively determine the predicted transaction behavior. Combining the predicted transaction date, preferred transaction type and predicted transaction amount, the predicted transaction behavior of the user on the predicted transaction date can be comprehensively determined. For example, it is predicted that the user will make a fund investment on the 10th of the next month, the transaction type is fund investment, and the predicted transaction amount is 6,000 yuan.

[0054] For example, the user's investment profile shows that he is a "regular fixed amount investor" who invests in funds at the end of each quarter, with an average investment amount of 10,000 yuan each time, and has stable investment returns in the past. Based on this profile, the predicted transaction date is the end of the next quarter, the transaction type is fund investment, and the predicted transaction amount is 10,000 yuan. One week before the predicted transaction date, financial institutions can push personalized fund investment advice to users through applications or emails, including recommended fund products and expected return analysis.

[0055] If the user's investment profile shows that he is a "market opportunity investor", he conducts stock transactions based on market dynamics, and the average transaction amount in the past year is 20,000 yuan. If the current market analysis shows that there is a favorable opportunity for stock investment, combined with the user's recent capital inflow, the predicted transaction date is the date when the market opportunity appears, the transaction type is stock trading, and the predicted transaction amount is 20,000 yuan or the amount adjusted according to the recent capital inflow. Before predicting the transaction date, financial institutions push market analysis and stock trading suggestions in the user's preferred method to attract users to trade.

[0056] This embodiment extracts key information such as investment cycle, investment behavior, investment products and investment results, and comprehensively analyzes and determines the predicted transaction date, transaction type and predicted transaction amount of the predicted transaction behavior, helping financial institutions to more accurately predict user needs, provide timely and personalized investment advice, and enhance user stickiness and satisfaction.

[0057] After determining the predicted transaction behavior, the products to be recommended can be determined based on the corresponding matching rules. Optionally, in the product recommendation method provided in the embodiment of the present application, determining the products to be recommended of the target financial institution based on the predicted transaction behavior includes: screening financial products that match both the transaction type and the predicted transaction amount from the target financial institution to obtain a group of financial products; obtaining historical financial products purchased by the user, and determining the similarity between each financial product in the group of financial products and the historical financial products; removing financial products whose similarity is less than a first similarity threshold from the group of financial products to obtain an updated group of financial products; and determining the updated group of financial products as the products to be recommended.

[0058] In some examples, the initial screening of financial products can be based on the transaction type and predicted transaction amount, and financial products that meet the user's needs can be screened out from the product library of the target financial institution. For example, if the transaction type is fund investment and the predicted transaction amount is 5,000 yuan, the system will screen out fund-related financial products and further refine them based on the amount matching conditions, such as screening out fund products with a minimum investment threshold of 5,000 yuan or less.

[0059] Obtain the records of financial products purchased by users in the past from historical financial product records, including information such as product type, investment amount, investment cycle, and return status, to establish a user's historical financial product database. Calculate the similarity between each product and the user's historical financial products. The similarity calculation can be based on a variety of factors, such as product type, investment threshold, risk level, expected rate of return, etc. Methods such as cosine similarity or similarity scoring based on machine learning can be used. Set a similarity threshold (first similarity threshold) and remove financial products with similarity to historical financial products below the threshold from a group of financial products to ensure that the recommended products have a high degree of match with the user's historical preferences and behavior patterns. Avoid recommending products that users may not be interested in or that are not suitable for their risk preferences.

[0060] For example, the user's investment profile shows that he is a "prudent investor" who has mainly invested in bond funds and money market funds in the past, with an average investment amount of 3,000 yuan. Based on this, the bond fund and money market fund categories with a minimum investment threshold of 3,000 yuan or less are screened out from the product library of the target financial institution. Then, the similarity is calculated with the historical financial products purchased by the user, and the first similarity threshold is set to 0.7. Products below this threshold are eliminated to obtain the final list of recommended products.

[0061] The user is an "active stock investor" with a high risk tolerance. He has frequently traded stocks in the past, with an average transaction amount of 10,000 yuan. Stocks and stock fund products are screened from the product library of the target financial institution, and the minimum transaction amount is 10,000 yuan or less. Then, the similarity is calculated with the user's historical stock transaction records, and the first similarity threshold is set to 0.6. Products with a similarity lower than 0.6 are eliminated to ensure that the recommended products are consistent with the user's past investment behavior and preferences, thereby improving the accuracy of the recommendation and the possibility of user purchase.

[0062] This embodiment further refines financial product recommendations by calculating and screening the similarity between the screened financial products and the user's historical financial products, improves the relevance and personalization of the recommendations, helps financial institutions meet user needs more accurately, and improves the success rate of product recommendations and user satisfaction.

[0063] After the product to be recommended is pushed to the user, the user's financial information is updated in a timely manner based on the recommendation result of the user for the recommended product. Optionally, in the product recommendation method provided in the embodiment of the present application, after the product to be recommended is pushed to the user's client, the method also includes: determining the recommendation result of the product to be recommended, wherein the recommendation result includes whether it has been purchased or not purchased; in the case where the recommendation result is that it has been purchased, determining the target financial product purchased by the user, and updating the user's financial information based on the purchase record of the target financial product.

[0064] In some examples, the recommendation results are determined by tracking user feedback and behavior. This can be achieved by monitoring the user's operations on the client, viewing product details, consulting customer service, completing purchases, and other behaviors. The recommendation results are divided into two states: "purchased" and "not purchased". If the user does purchase the recommended financial product, identify the specific financial product purchased by the user (target financial product), as well as the purchase date, amount, product type and other detailed information. Update the user's financial information based on the record of the target financial product purchased by the user. The updated content may include but is not limited to: update of total assets, including the purchase amount in the user's total assets. Update of investment portfolio, adding the target financial product to the user's current investment portfolio. Update of risk assessment, adjusting the user's risk assessment according to the risk characteristics of the newly purchased product. Update of investment cycle, if the target product has a certain investment cycle, update the user's financial management cycle information. Update of income forecast, adjusting the user's income forecast based on the expected income of the product.

[0065] For example, when the recommendation results show that the user has purchased the recommended bond fund product, the user's financial information is immediately updated, including the purchase amount being included in the total assets, the bond fund being added to the investment portfolio, and the user's investment profile being adjusted according to the risk level and investment cycle of the bond fund. In addition, the date and amount of the purchase behavior are recorded as input to subsequent recommendation and prediction algorithms.

[0066] If a user makes a purchase after receiving a stock recommendation, not only will the user’s financial information be updated, including the type, quantity, purchase date and amount of the stock, but the purchase behavior will also be considered positive feedback on the effectiveness of the recommendation system, further optimizing the recommendation algorithm and improving the accuracy and personalization of future recommendations.

[0067] This embodiment forms a data-driven closed-loop system by tracking and determining the recommendation results and updating the user's financial information based on the user's purchasing behavior. It improves the efficiency and accuracy of the recommendation system, dynamically adjusts the user's financial profile, and ensures that the financial institution's recommendations are always in sync with the user's needs and behaviors.

[0068] If the user has not purchased the recommended financial product, it is also necessary to detect whether the user has purchased other financial products, and then update the financial information to ensure the timeliness of the user information. Optionally, in the product recommendation method provided in the embodiment of the present application, after determining the recommendation result of the product to be recommended, the method also includes: if the recommendation result is not purchased, detecting whether the user has purchased other financial products, wherein the other financial products are financial products other than the product to be recommended; if the user is not detected to have purchased other financial products, updating the user's financial information at preset time intervals to obtain updated financial information; if the similarity between the updated financial information and the financial information within a preset period is less than or equal to a second similarity threshold, determining an updated product to be recommended based on the updated financial information, and pushing the updated product to be recommended to the user's client.

[0069] In some examples, if the user does not purchase the recommended financial product, the user's behavior is continuously monitored to check whether the user has selected other financial products in addition to the recommendation. If the user is not detected to purchase the recommended product or other financial products, the user's financial information is automatically updated at preset time intervals (such as weekly, monthly, etc.) to reflect changes in their assets, transactions, investment portfolios, etc. The updated financial information can be based on the user's latest financial data, market trends, and possible personal financial events (such as salary payments, bonuses, inheritance, etc.).

[0070] Compare the updated financial information with the financial information of the previous update cycle and calculate their similarity. If the similarity is less than or equal to the set second similarity threshold, it means that the user's financial situation has changed significantly, and his investment preferences and behavior patterns have changed. Based on the updated financial information and similarity analysis results, re-evaluate the user's investment profile and potential needs, and determine the updated products to be recommended. Push the updated products to be recommended to the user through the user's client.

[0071] For example, after a group of bond products are recommended, the user does not make any purchases. It is detected that the user has not purchased any financial products within two weeks after the recommendation, so the user's financial information is updated according to the preset monthly update cycle. By comparison, if the similarity between the updated financial information and the previous month is less than the second similarity threshold (such as 0.8), the user's investment needs are re-analyzed based on the latest financial information, and a group of financial products that better match the user's investment profile are recommended, and pushed through the user's preferred channels.

[0072] This embodiment continuously optimizes financial product recommendations by detecting user behavior, regularly updating financial information, and performing similarity analysis to ensure that the recommendations are consistent with the user's financial status and investment preferences.

[0073] If the user has purchased other financial products, it is necessary to timely update the user's financial information based on the other financial products. Optionally, in the product recommendation method provided in the embodiment of the present application, after determining the recommendation result of the product to be recommended, the method also includes: if the recommendation result is not purchased, detecting whether the user has purchased other financial products, wherein the other financial products are financial products other than the product to be recommended; if it is detected that the user has purchased other financial products, determining whether the other financial products are products of the target financial institution; if the other financial products are products of the target financial institution, updating the user's financial information based on the purchase records of the other financial products; if the other financial products are not products of the target financial institution, obtaining the financial products in the target financial institution with the greatest similarity to the other financial products to obtain similar financial products; and updating the user's financial information based on the product information of the similar financial products.

[0074] In some examples, if the user does not purchase the recommended financial product, detect whether the user has purchased the financial product through other channels. This can be achieved by analyzing the user's transaction records at third-party financial institutions, public market data, and investment behaviors actively reported by the user. Determine whether other financial products purchased by the user belong to the target financial institution. If so, integrate the purchase record of the product into the user's financial information, and update the data such as the investment portfolio, total assets, and risk assessment.

[0075] If the other financial products purchased by the user do not belong to the target financial institution, find the financial products that are most similar to the products purchased by the user from the product library of the target financial institution. Update the user's financial information based on the similar financial product information found in the target financial institution. For example, if the user purchases a high-risk stock product, and the target financial institution provides stock funds with similar risk levels, then the update of the financial information should reflect the user's interest in high-risk products, adjust the risk preference in its investment profile, and update the corresponding product category information in the investment portfolio.

[0076] For example, a financial institution recommends a set of time deposit products to a user, but the user does not purchase them. Later, it is detected that the user has purchased a high-risk stock product from another financial institution. First, determine whether this is a product of the target financial institution. If not, find stock fund products with similar risk levels and expected returns from the product library of the target financial institution through similarity calculation, and update the user's financial information based on the information of the similar financial products, including raising the risk preference level and adjusting the product ratio in the investment portfolio.

[0077] This embodiment detects other financial products purchased by users, determines their sources, and updates the user's financial information based on similar financial product information of the target financial institution, forming a more flexible and adaptive recommendation and information update mechanism. It can more accurately understand the user's behavior and preferences, adjust product recommendations in a timely manner, and improve user satisfaction and the possibility of product purchase.

[0078] In order to determine the user's investment behavior pattern and predict the user's possible transaction behavior, it is necessary to train a transaction model. Optionally, in the product recommendation method provided in the embodiment of the present application, the transaction model is trained in the following manner: obtaining the transaction records of the target financial institution, and extracting each user's historical financial management information and historical investment portrait from the transaction records; determining each user's historical financial management information and historical investment portrait as a group of training samples to obtain multiple groups of training samples; training a machine learning model based on the multiple groups of training samples to obtain a transaction model.

[0079] In some examples, first, the user's transaction records are obtained from the database of the target financial institution. These records may contain rich historical financial information and investment behavior data. Historical financial information may include indicators such as total assets, investment portfolio, transaction frequency, and risk preference. Historical investment profiles may include characteristics such as the user's investment behavior pattern, preferred product type, and investment cycle. For each user, historical financial information and historical investment profiles are extracted from their transaction records to form a set of training samples. The training samples of all users are aggregated to obtain a large data set containing historical data of multiple users, that is, multiple sets of training samples. When processing these training samples, pre-processing steps such as data cleaning, feature selection, and outlier processing can be performed to ensure data quality and improve the training effect of subsequent machine learning models.

[0080] Based on multiple sets of training samples, select appropriate machine learning algorithms (such as decision trees, random forests, support vector machines, neural networks, etc.) for training and establish trading models. The goal of the trading model is to learn the user's investment behavior patterns, predict future trading possibilities and amounts, and recommend financial products that best meet user needs and preferences. During the model training process, cross-validation, grid search and other technologies can be used to optimize parameters to improve the model's generalization ability and prediction accuracy.

[0081] For example, if a customer has a habit of investing in funds, this behavior can characterize the customer's risk preference for fixed investment funds, the fixed investment amount, and the fixed investment ratio in the account deposit, and a trading model can be established. When the customer's account has income (such as payday, or stock income redemption), the trading model can be used to determine the customer's preferred fund products or better combination products that better suit the user's investment habits.

[0082] This embodiment constructs training samples through the transaction records of users of the target financial institution, and trains the transaction model through machine learning technology to achieve accurate prediction of users' future transaction behaviors and personalized product recommendations.

[0083] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0084] Example 2

[0085] The embodiment of the present application also provides a product recommendation device. It should be noted that the product recommendation device of the embodiment of the present application can be used to execute the product recommendation method provided by the embodiment of the present application. The product recommendation device provided by the embodiment of the present application is introduced below.

[0086] According to an embodiment of the present application, a device for implementing the above product recommendation method is also provided. Figure 3 is a schematic diagram of a product recommendation device provided according to an embodiment of the present application, such as Figure 3 As shown, the device comprises:

[0087] The acquisition unit 301 is used to acquire the financial information of the user within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information and investment information;

[0088] An input unit 302 is used to input the financial information into the transaction model to obtain the user's investment profile, wherein the transaction model is trained by multiple sets of training samples, each set of training samples includes historical financial information and historical investment profiles, and the investment profile is used to characterize the user's investment habits;

[0089] A first determining unit 303 is used to determine the user's predicted transaction behavior based on the investment profile, wherein the predicted transaction behavior includes at least one of the following: deposit, mortgage, stock transaction and fund transaction;

[0090] The second determining unit 304 is used to determine the products to be recommended of the target financial institution based on the predicted transaction behavior, and push the products to be recommended to the user's client.

[0091] The product recommendation device provided in the embodiment of the present application obtains the financial information of the user within a preset period through the acquisition unit 301, wherein the financial information includes at least one of the following: identity information, transaction information and investment information; the input unit 302 inputs the financial information into the transaction model to obtain the investment portrait of the user, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the investment habits of the user; the first determination unit 303 determines the user's predicted transaction behavior based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposits, mortgages, stock transactions and fund transactions; the second determination unit 304 determines the target financial institution's to-be-recommended products based on the predicted transaction behavior, and pushes the to-be-recommended products to the user's client, determines the user's predicted transaction behavior by analyzing the user's financial information, and then determines the target financial institution's to-be-recommended products based on the predicted transaction behavior, and recommends the product to the user, thereby achieving the purpose of personalized recommendation of financial products based on the user's predicted transaction behavior, thereby achieving the technical effect of improving the matching degree between the recommended products and the user's product purchase tendency, and then solving the technical problem of the mismatch between the recommended products and the user's product purchase tendency.

[0092] Optionally, in the product recommendation device provided in the embodiment of the present application, the first determination unit 303 includes: an extraction module, used to extract the user's investment cycle, investment behavior, investment products and investment results from the investment portrait; a first determination module, used to determine the predicted transaction date of the predicted transaction behavior based on the investment cycle, determine the transaction type preferred by the user based on the investment behavior, and determine the predicted transaction amount based on the investment product and investment results; a second determination module, used to determine the predicted transaction behavior on the predicted transaction date through the transaction type and the predicted transaction amount.

[0093] Optionally, in the product recommendation device provided in the embodiment of the present application, the second determination unit 304 includes: a screening module, used to screen financial products that match both the transaction type and the predicted transaction amount from the target financial institution to obtain a group of financial products; an acquisition module, used to obtain historical financial products purchased by the user, and determine the similarity between each financial product in a group of financial products and the historical financial products; a removal module, used to remove financial products whose similarity is less than a first similarity threshold from a group of financial products to obtain an updated group of financial products; and a third determination module, used to determine the updated group of financial products as products to be recommended.

[0094] Optionally, in the product recommendation device provided in the embodiment of the present application, the device also includes: a third determination unit, used to determine the recommendation result of the product to be recommended, wherein the recommendation result includes purchased or not purchased; a first update unit, used to determine the target financial product purchased by the user when the recommendation result is purchased, and update the user's financial information based on the purchase record of the target financial product.

[0095] Optionally, in the product recommendation device provided in the embodiment of the present application, the device also includes: a first detection unit, used to detect whether the user purchases other financial products when the recommendation result is not purchased, wherein the other financial products are financial products other than the product to be recommended; a second updating unit, used to update the user's financial information at preset time intervals to obtain updated financial information when the user is not detected to have purchased other financial products; and a push unit, used to determine an updated product to be recommended based on the updated financial information when the similarity between the updated financial information and the financial information within a preset period is less than or equal to a second similarity threshold, and push the updated product to be recommended to the user's client.

[0096] Optionally, in the product recommendation device provided in the embodiment of the present application, the device also includes: a second detection unit, used to detect whether the user purchases other financial products when the recommendation result is not purchased, wherein the other financial products are financial products other than the product to be recommended; a judgment unit, used to judge whether the other financial products are products of the target financial institution when it is detected that the user purchases other financial products; a third updating unit, used to update the user's financial information based on the purchase record of the other financial products when the other financial products are products of the target financial institution; a product acquisition unit, used to acquire the financial products with the greatest similarity to other financial products in the target financial institution when the other financial products are not products of the target financial institution, and obtain similar financial products; a fourth updating unit, used to update the user's financial information based on the product information of the similar financial products.

[0097] Optionally, in the product recommendation device provided in the embodiment of the present application, the device also includes: an extraction unit, used to obtain transaction records of the target financial institution, and extract each user's historical financial management information and historical investment portrait from the transaction records; a fourth determination unit, used to determine each user's historical financial management information and historical investment portrait as a group of training samples to obtain multiple groups of training samples; a training unit, used to train a machine learning model based on multiple groups of training samples to obtain a transaction model.

[0098] It should be noted that the acquisition unit 301, input unit 302, first determination unit 303 and second determination unit 304 correspond to steps S201 to S204 in Example 1, and the four units and corresponding steps implement the same examples and application scenarios, but are not limited to the contents disclosed in the above-mentioned Example 1. It should be noted that the above-mentioned modules or units may be hardware components or software components stored in a memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n), and the above-mentioned modules or units may also be part of a device and may be run in the computer terminal 10 provided in Example 1.

[0099] Example 3

[0100] An embodiment of the present application may provide an electronic device, Figure 4 is a structural block diagram of an electronic device according to an embodiment of the present application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (only one is shown) processor 402, memory 404, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0101] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0102] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain the user's financial information within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information and investment information; input the financial information into the transaction model to obtain the user's investment portrait, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the user's investment habits; determine the user's predicted transaction behavior based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposits, mortgages, stock transactions and fund transactions; determine the target financial institution's products to be recommended based on the predicted transaction behavior, and push the products to be recommended to the user's client.

[0103] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: extract the user's investment cycle, investment behavior, investment products and investment results from the investment portrait; determine the predicted transaction date of the predicted transaction behavior based on the investment cycle, determine the transaction type preferred by the user based on the investment behavior, and determine the predicted transaction amount based on the investment products and investment results; determine the predicted transaction behavior on the predicted transaction date through the transaction type and the predicted transaction amount.

[0104] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: screen financial products that match the transaction type and predicted transaction amount from the target financial institution to obtain a group of financial products; obtain historical financial products purchased by the user, and determine the similarity between each financial product in the group of financial products and the historical financial products; remove financial products whose similarity is less than a first similarity threshold from the group of financial products to obtain an updated group of financial products; and determine the updated group of financial products as products to be recommended.

[0105] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: determine the recommendation result of the product to be recommended, wherein the recommendation result includes whether it has been purchased or not; if the recommendation result is purchased, determine the target financial product purchased by the user, and update the user's financial information based on the purchase record of the target financial product.

[0106] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: when the recommendation result is not purchased, detect whether the user purchases other financial products, where other financial products are financial products other than the product to be recommended; when the user is not detected to have purchased other financial products, update the user's financial information at preset time intervals to obtain updated financial information; when the similarity between the updated financial information and the financial information within a preset period is less than or equal to a second similarity threshold, determine an updated product to be recommended based on the updated financial information, and push the updated product to be recommended to the user's client.

[0107] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: when the recommendation result is not purchased, detect whether the user purchases other financial products, where other financial products are financial products other than the product to be recommended; when it is detected that the user purchases other financial products, determine whether the other financial products are products of the target financial institution; when the other financial products are products of the target financial institution, update the user's financial information based on the purchase records of the other financial products; when the other financial products are not products of the target financial institution, obtain the financial products in the target financial institution with the greatest similarity to other financial products to obtain similar financial products; and update the user's financial information based on the product information of the similar financial products.

[0108] The processor can also call the information and applications stored in the memory through the transmission device to perform the following steps: obtain the transaction records of the target financial institution, and extract the historical financial information and historical investment portrait of each user from the transaction records; determine the historical financial information and historical investment portrait of each user as a set of training samples to obtain multiple sets of training samples; train the machine learning model based on the multiple sets of training samples to obtain a transaction model.

[0109] By adopting the embodiment of the present application, a method is provided for obtaining the financial information of a user within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information and investment information; inputting the financial information into a transaction model to obtain the investment portrait of the user, wherein the transaction model is trained by multiple groups of training samples, each group of training samples includes historical financial information and historical investment portraits, and the investment portrait is used to characterize the investment habits of the user; determining the predicted transaction behavior of the user based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposits, mortgages, stock transactions and fund transactions; determining the target financial institution's to-be-recommended products based on the predicted transaction behavior, and pushing the to-be-recommended products to the user's client. By analyzing the user's financial information to determine the user's predicted transaction behavior, and then determining the target financial institution's to-be-recommended products based on the predicted transaction behavior, and recommending the products to the user, the purpose of personalized recommendation of financial products based on the user's predicted transaction behavior is achieved, thereby achieving the technical effect of improving the matching degree between the recommended products and the user's product purchase tendency, and then solving the technical problem of the mismatch between the recommended products and the user's product purchase tendency.

[0110] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the electronic device may also be a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), a PAD or other terminal device. Figure 4 The structure of the electronic device is not limited. Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 4 Different configurations are shown.

[0111] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0112] Example 4

[0113] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the product recommendation method provided in the first embodiment.

[0114] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0115] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the product recommendation method.

[0116] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0117] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0118] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0121] If the integrated unit is implemented in the form of 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 the present 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, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0122] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A product recommendation method, characterized in that: include: Acquire the user's financial information within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information, and investment information; Inputting the financial information into a transaction model to obtain an investment profile of the user, wherein the transaction model is trained by multiple sets of training samples, each set of training samples includes historical financial information and historical investment profiles, and the investment profile is used to characterize the investment habits of the user; Determining the predicted transaction behavior of the user based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposit, mortgage, stock transaction and fund transaction; The recommended products of the target financial institution are determined based on the predicted transaction behavior, and the recommended products are pushed to the client of the user.

2. The method according to claim 1, characterized in that: Determining the predicted transaction behavior of the user based on the investment profile includes: Extracting the user's investment cycle, investment behavior, investment products and investment results from the investment portrait; Determining a predicted transaction date of the predicted transaction behavior based on the investment cycle, determining a transaction type preferred by the user based on the investment behavior, and determining a predicted transaction amount based on the investment product and the investment result; The predicted transaction behavior on the predicted transaction date is determined by the transaction type and the predicted transaction amount.

3. The method according to claim 2, characterized in that The products to be recommended by the target financial institution determined based on the predicted transaction behavior include: Selecting financial products that match both the transaction type and the predicted transaction amount from the target financial institution to obtain a group of financial products; Acquire historical financial products purchased by the user, and determine the similarity between each financial product in the group of financial products and the historical financial products; Eliminate financial products whose similarity is less than a first similarity threshold from the group of financial products to obtain an updated group of financial products; The updated group of financial products is determined as the products to be recommended.

4. The method according to claim 1, characterized in that After pushing the product to be recommended to the client of the user, the method further includes: Determine a recommendation result of the product to be recommended, wherein the recommendation result includes whether the product has been purchased or not purchased; In the case that the recommendation result is that the product has been purchased, the target financial product purchased by the user is determined, and the financial management information of the user is updated based on the purchase record of the target financial product.

5. The method according to claim 4, characterized in that After determining the recommendation result of the product to be recommended, the method further includes: In the case where the recommendation result is not purchased, detecting whether the user purchases other financial products, wherein the other financial products are financial products other than the product to be recommended; When it is detected that the user has not purchased the other financial products, updating the financial information of the user at a preset time interval to obtain updated financial information; When the similarity between the updated financial information and the financial information within the preset period is less than or equal to a second similarity threshold, an updated product to be recommended is determined based on the updated financial information, and the updated product to be recommended is pushed to the user's client.

6. The method according to claim 4, characterized in that After determining the recommendation result of the product to be recommended, the method further includes: In the case where the recommendation result is not purchased, detecting whether the user purchases other financial products, wherein the other financial products are financial products other than the product to be recommended; In the case of detecting that the user purchases the other financial product, determining whether the other financial product is a product of the target financial institution; If the other financial product is a product of the target financial institution, updating the financial management information of the user based on the purchase record of the other financial product; In the case where the other financial product is not a product of the target financial institution, obtaining a financial product in the target financial institution that has the greatest similarity to the other financial product to obtain a similar financial product; The financial management information of the user is updated based on the product information of the similar financial products.

7. The method according to claim 1, characterized in that The transaction model is trained in the following way: Obtaining the transaction records of the target financial institution, and extracting each user's historical financial information and historical investment profile from the transaction records; Determine each user's historical financial information and historical investment profile as a set of training samples, and obtain multiple sets of training samples; A machine learning model is trained based on the multiple groups of training samples to obtain the transaction model.

8. A product recommendation device, characterized in that: include: An acquisition unit, configured to acquire the user's financial information within a preset period, wherein the financial information includes at least one of the following: identity information, transaction information, and investment information; An input unit, used to input the financial information into a transaction model to obtain an investment profile of the user, wherein the transaction model is trained by multiple sets of training samples, each set of training samples includes historical financial information and historical investment profiles, and the investment profile is used to characterize the investment habits of the user; A first determining unit, configured to determine a predicted transaction behavior of the user based on the investment portrait, wherein the predicted transaction behavior includes at least one of the following: deposit, mortgage, stock transaction, and fund transaction; The second determining unit is used to determine the products to be recommended of the target financial institution based on the predicted transaction behavior, and push the products to be recommended to the client of the user.

9. An electronic device, characterized in that: include: A memory storing an executable program; A processor is used to run the program, wherein the program, when running, executes the product recommendation method described in any one of claims 1 to 7.

10. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the product recommendation method described in any one of claims 1 to 7 are implemented.

Images