Information recommendation method, information recommendation device, electronic device and readable storage medium

Abstract

The disclosed embodiments provide an information recommendation method, an information recommendation device, an electronic device, and a readable storage medium. The method includes: obtaining target financial behavior information of a target user, wherein the target financial behavior information of the target user includes at least one of the following: user flow information, user risk assessment information, and financial product information of the target user; inputting the target financial behavior information of the target user into an information recommendation model, and outputting a recommendation score for each candidate financial information in a plurality of candidate financial information; determining the target financial information from the plurality of candidate financial information according to the plurality of recommendation scores; and recommending the target financial information to the target user. The information recommendation method and device provided by the disclosed embodiments can be used in the financial field or other fields.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the field of artificial intelligence technology, and more specifically, to an information recommendation method, an information recommendation device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the rapid development of the financial industry, financial institutions currently recommend some financial information to be promoted to various users.

[0003] In the process of realizing the concept of the present disclosure, the inventors found that there are at least the following problems in the related art: it is difficult to achieve personalized financial information recommendations using the related art. Summary of the invention

[0004] In view of this, embodiments of the present disclosure provide an information recommendation method, an information recommendation device, an electronic device, a computer-readable storage medium, and a computer program product.

[0005] One aspect of an embodiment of the present disclosure provides an information recommendation method, comprising: obtaining target financial behavior information of a target user, wherein the target financial behavior information of the target user includes at least one of the following: user flow information, user risk assessment information and financial product information of the target user; inputting the target financial behavior information of the target user into an information recommendation model, and outputting a recommendation score for each candidate financial information among multiple candidate financial information; determining target financial information from the multiple candidate financial information according to the multiple recommendation scores; and recommending the target financial information to the target user.

[0006] According to an embodiment of the present disclosure, the above-mentioned acquisition of target financial behavior information of the target user includes: acquiring the original financial behavior information of the above-mentioned target user; determining user sensitive information related to user privacy in the above-mentioned original financial behavior information; processing the above-mentioned user sensitive information using a data desensitizing method to obtain processed user sensitive information; and determining the above-mentioned target financial behavior information based on the above-mentioned original financial behavior information and the above-mentioned processed user sensitive information.

[0007] According to an embodiment of the present disclosure, the above-mentioned information recommendation model includes a click-through rate model and a revenue model, wherein the above-mentioned click-through rate model is obtained by jointly training a first long short-term memory neural network model, a first deep neural network model and a first neural network model, and the above-mentioned revenue model is obtained by jointly training a second long short-term memory neural network model, a second deep neural network model and a second neural network model.

[0008] According to an embodiment of the present disclosure, the target financial behavior information of the target user is input into the information recommendation model, and a recommendation score for each candidate financial information among multiple candidate financial information is output, including: inputting the target financial behavior information of the target user into the click-through rate model, and outputting a click-through rate for each candidate financial information among the multiple candidate financial information; inputting the target financial behavior information of the target user into the benefit model, and outputting a benefit for each candidate financial information among the multiple candidate financial information; and determining a recommendation score for the candidate financial information according to the click-through rate and benefit of each candidate financial information among the multiple candidate financial information.

[0009] According to an embodiment of the present disclosure, the click-through rate model is obtained by jointly training a first long short-term memory neural network model, a first deep neural network model and a first neural network model, including: obtaining a first training sample, wherein the first training sample includes multiple positive samples and multiple negative samples, each of the positive samples includes the target financial behavior information of the sample user and the actual click-through rate of the sample user for the sample product, each of the negative samples includes the target training financial behavior information of the sample user and the actual display information of the sample user for the sample product, and the target training financial behavior information of the sample user includes the user flow information, user risk assessment information and financial product information of the sample user; inputting the user flow information of the sample user included in each of the multiple positive samples into the first long short-term memory neural network model, and outputting a first vector; inputting the user risk assessment information and financial product information of the sample user included in each of the multiple positive samples into the first deep neural network model, and outputting a second vector; and inputting the second vector and The above-mentioned second vectors are spliced ​​to obtain a third vector; the above-mentioned third vector is input into the above-mentioned first neural network model, and the predicted click rate of the above-mentioned sample users included in the above-mentioned positive samples for the above-mentioned sample products is output; the user flow information of the sample users included in each negative sample of the above-mentioned multiple negative samples is input into the above-mentioned first long short-term memory neural network model, and a fourth vector is output; the user risk assessment information and financial product information of the sample users included in each negative sample of the above-mentioned multiple negative samples are input into the above-mentioned first deep neural network model, and a fifth vector is output; the above-mentioned fourth vector and the above-mentioned fifth vector are spliced ​​to obtain a sixth vector; the above-mentioned sixth vector is input into the above-mentioned first neural network model, and the predicted display information of the above-mentioned sample users included in the above-mentioned negative samples for the above-mentioned sample products is output; and the above-mentioned first long short-term memory neural network model, the above-mentioned first deep neural network model and the above-mentioned first neural network model are trained according to the multiple above-mentioned real click rates, the multiple above-mentioned predicted click rates, the multiple above-mentioned real display information and the multiple above-mentioned predicted display information to obtain the above-mentioned click rate model.

[0010] According to an embodiment of the present disclosure, the above-mentioned profit model is obtained by joint training using a second long short-term memory neural network model, a second deep neural network model and a second neural network model, including: obtaining a second training sample, wherein the above-mentioned second training sample includes multiple training groups, each of the above-mentioned training groups includes the target financial behavior information of the above-mentioned sample users and the actual benefits of the above-mentioned sample users for the sample products, and the target training financial behavior information of the above-mentioned sample users includes the user flow information, user risk assessment information and financial product information of the above-mentioned sample users; inputting the user flow information of the sample users included in each of the above-mentioned multiple training groups into the above-mentioned second long short-term memory neural network model, and outputting a seventh vector; inputting the user risk assessment information and financial product information of the sample users included in each of the above-mentioned multiple training groups into the above-mentioned second deep neural network model, and outputting an eighth vector; concatenating the above-mentioned seventh vector and the above-mentioned eighth vector to obtain a ninth vector; inputting the above-mentioned ninth vector into the above-mentioned second neural network model, and outputting the predicted benefits of the above-mentioned sample users included in the above-mentioned training group for the above-mentioned sample products; and training the above-mentioned second long short-term memory neural network model, the above-mentioned second deep neural network model and the above-mentioned second neural network model according to the multiple above-mentioned actual benefits and the multiple above-mentioned predicted benefits to obtain the above-mentioned profit model.

[0011] According to an embodiment of the present disclosure, determining the target financial information from the multiple candidate financial information based on the multiple recommendation scores includes: sorting the multiple recommendation scores to obtain a sorting result; and determining the target financial information from the multiple candidate financial information based on the sorting result.

[0012] According to an embodiment of the present disclosure, the above-mentioned user transaction information includes at least one of the following: transaction time, transaction amount and transaction nature of transactions within a preset time period; the above-mentioned financial product information includes at least one of the following: risk level of the financial product, fund manager, asset size, buying cycle and redemption cycle.

[0013] According to an embodiment of the present disclosure, the above-mentioned information recommendation model is trained using a federated learning framework.

[0014] Another aspect of the disclosed embodiment provides an information recommendation device, including an acquisition module, an output module, a determination module and a recommendation module. The acquisition module is used to acquire the target financial behavior information of the target user, wherein the target financial behavior information of the target user includes at least one of the following: the user flow information, user risk assessment information and financial product information of the target user; the output module is used to input the target financial behavior information of the target user into the information recommendation model, and output the recommendation score for each candidate financial information in multiple candidate financial information; the determination module is used to determine the target financial information from the multiple candidate financial information according to the multiple recommendation scores; and the recommendation module is used to recommend the target financial information to the target user.

[0015] Another aspect of an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described above.

[0016] Another aspect of an embodiment of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which are used to implement the method described above when executed.

[0017] Another aspect of an embodiment of the present disclosure provides a computer program product, wherein the computer program product includes computer executable instructions, and the instructions are used to implement the above method when executed.

[0018] According to the embodiments of the present disclosure, the target financial behavior of the target user is processed by using the information recommendation model to obtain the recommendation score of each candidate financial information in multiple candidate financial information, and the target financial information recommended to the target user is determined based on the multiple recommendation scores. Since the target financial behavior information of the target user can reflect the personalized financial information of the user, the target financial behavior information of the target user is processed by using the information recommendation model to obtain the recommendation score of each candidate financial information in multiple candidate financial information, and the target financial information recommended to the target user is determined based on the multiple recommendation scores, thereby realizing personalized recommendation, and the target financial behavior information is processed by using the information recommendation model to improve the accuracy of the recommendation, thereby meeting the needs of the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0020] Figure 1 An exemplary system architecture 100 to which the information recommendation method according to an embodiment of the present disclosure can be applied is schematically shown;

[0021] Figure 2 The flowchart of the information recommendation method 200 according to the embodiment of the present disclosure is schematically shown;

[0022] Figure 3 A schematic diagram schematically shows a training process of an information recommendation model according to an embodiment of the present disclosure;

[0023] Figure 4 A block diagram schematically shows an information recommendation device 400 according to an embodiment of the present disclosure; and

[0024] Figure 5 A block diagram of an electronic device 500 suitable for implementing the method described above according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0025] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0026] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0027] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0028] In the case of using expressions such as "at least one of A, B, and C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.). In the case of using expressions such as "at least one of A, B, or C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, or C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0029] In the process of realizing the concept of the present disclosure, the inventors found that most of the related technologies use manual methods to recommend financial information, and recommend the same financial information to different target users, making it difficult to achieve personalized recommendations. In addition, the methods of using models to recommend financial information in the related technologies are often based on static user information, ignoring the changes in user demand for financial information, affecting the accuracy of user recommendations, and thus making it difficult to meet user needs.

[0030] In order to solve the problems existing in the related art, the inventor proposes a scheme for determining the target financial information to be recommended by processing the target financial behavior information of the target user using an information recommendation model, wherein the target financial behavior information of the target user includes at least one of the target user's user flow information, user risk assessment information, and financial product information. Since the target financial behavior information of the target user can reflect the personalized financial information of the user, the target financial behavior information of the target user is processed by the information recommendation model to determine the target financial information recommended to the target user, thereby realizing personalized recommendation, and the target financial behavior information is processed by the information recommendation model to improve the accuracy of the recommendation, thereby meeting the needs of the user.

[0031] Specifically, the embodiments of the present disclosure provide an information recommendation method, an information recommendation device, an electronic device, and a readable storage medium. The method includes an acquisition process, an output process, a determination process, and a recommendation process. In the acquisition process, the target financial behavior information of the target user is acquired, and the target financial behavior information of the target user includes at least one of the following: the user flow information, user risk assessment information, and financial product information of the target user; in the output process, the target financial behavior information of the target user is input into the information recommendation model, and a recommendation score for each candidate financial information in a plurality of candidate financial information is output; in the determination process, the target financial information is determined from a plurality of candidate financial information according to a plurality of recommendation scores; in the recommendation process, the target financial information is recommended to the target user.

[0032] It should be noted that the information recommendation method and device provided by the embodiments of the present disclosure can be applied to the financial field, for example, for bank staff to recommend financial products to customers; in addition, the information recommendation method and device provided by the embodiments of the present disclosure can also be applied to other fields outside the financial field, such as in the field of Internet technology, where the information provider can push appropriate content based on the user's behavior to improve the user retention rate of the information provider. The application field of the information recommendation method and device provided by the embodiments of the present disclosure is not limited.

[0033] Figure 1 The exemplary system architecture 100 to which the information recommendation method according to the embodiment of the present disclosure can be applied is schematically shown. It should be noted that: Figure 1 What is shown is merely an example of a system architecture to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0034] like Figure 1 As shown, the system architecture 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0035] The account can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 101, 102, 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients and / or social platform software, etc. (only as examples).

[0036] The terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0037] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by the account using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as the account request, and feed back the processing results (such as the web page, information, or data obtained or generated according to the account request) to the terminal device.

[0038] It should be noted that the information recommendation method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the information recommendation device provided in the embodiment of the present disclosure can generally be set in the server 105. The information recommendation method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Accordingly, the information recommendation device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Alternatively, the information recommendation method provided in the embodiment of the present disclosure can be executed by the terminal devices 101, 102, or 103, or can also be executed by other terminal devices different from the terminal devices 101, 102, or 103. Accordingly, the information recommendation device provided in the embodiment of the present disclosure can also be set in the terminal devices 101, 102, or 103, or can be set in other terminal devices different from the terminal devices 101, 102, or 103.

[0039] For example, the target financial behavior information of the target user may be originally stored in any one of the terminal devices 101, 102, or 103 (for example, the terminal device 101, but not limited thereto), or stored in an external storage device and may be imported into the terminal device 101. Then, the terminal device 101 may locally execute the information recommendation method provided by the embodiment of the present disclosure, or send the target financial behavior information to other terminal devices, servers, or server clusters, and the other terminal devices, servers, or server clusters that receive the target financial behavior information may execute the information recommendation method provided by the embodiment of the present disclosure.

[0040] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0041] Figure 2 The flowchart of the information recommendation method 200 according to the embodiment of the present disclosure is schematically shown.

[0042] like Figure 2 As shown, the method 200 includes operations S210 to S240.

[0043] In operation S210, target financial behavior information of a target user is acquired.

[0044] According to an embodiment of the present disclosure, the target financial behavior information of the target user may refer to the information generated by the target user in daily transaction behaviors and recorded by the database, and may include a combination of one or more of user transaction information, user risk assessment information and financial product information.

[0045] According to an embodiment of the present disclosure, a privacy information use agreement may be signed with a target user, and after obtaining the target user's consent, the target financial behavior information may be extracted from a database.

[0046] It should be noted that the acquisition, storage and application of user personal information involved in the technical solution of the present disclosure are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0047] In operation S220 , the target financial behavior information of the target user is input into the information recommendation model, and a recommendation score for each of the plurality of candidate financial information is output.

[0048] According to an embodiment of the present disclosure, the information recommendation model can be based on an existing model architecture, using the historical financial behavior information of multiple users as training samples, and using the actual financial information selection of multiple users as labels for training. The number of neurons in the output layer of the information recommendation model is greater than the number of candidate financial information, so as to achieve the scalability of the information recommendation model.

[0049] According to the embodiments of the present disclosure, financial information may refer to specific financial products, such as fund products that a fund issuer entrusts a bank to sell; on the other hand, financial information may also refer to financial services, such as investment service recommendations for credit, stocks, etc.; in addition, financial information may also include finance-related news, popular science articles, product introduction articles, videos, etc.

[0050] According to an embodiment of the present disclosure, the recommendation score of the candidate financial information may indicate the degree of matching between the target user and the candidate financial information based on the analysis of the financial behavior information. The higher the recommendation score, the greater the probability that the target user will select the candidate financial information.

[0051] In operation S230, target financial information is determined from among a plurality of candidate financial information according to the plurality of recommendation scores.

[0052] In operation S240, target financial information is recommended to a target user.

[0053] According to the embodiments of the present disclosure, the target financial behavior of the target user is processed by using the information recommendation model to obtain the recommendation score of each candidate financial information in multiple candidate financial information, and the target financial information recommended to the target user is determined based on the multiple recommendation scores. Since the target financial behavior information of the target user can reflect the personalized financial information of the user, the target financial behavior information of the target user is processed by using the information recommendation model to obtain the recommendation score of each candidate financial information in multiple candidate financial information, and the target financial information recommended to the target user is determined based on the multiple recommendation scores, thereby realizing personalized recommendation, and the target financial behavior information is processed by using the information recommendation model to improve the accuracy of the recommendation, thereby meeting the needs of the user.

[0054] Reference below Figure 3 , combined with specific embodiments Figure 2 The method shown is further explained.

[0055] According to an embodiment of the present disclosure, user transaction information includes information generated by the target user's financial behavior through various channels within a preset time range. Specifically, the user transaction information may include information such as transaction time, transaction amount, and transaction nature. For example, the user transaction information may include information on the time and amount of stock transactions performed by the user within 180 days.

[0056] According to the embodiments of the present disclosure, the user risk assessment information can be determined by the product provider based on the historical financial behavior information of the target user. For example, the bank can determine the target user's risk resistance by querying the target user's credit information, thereby determining the target user's risk assessment information. In addition, the user risk assessment information can also be obtained by the user through self-testing means. For example, the target user completes a questionnaire test based on his or her personality tendencies to obtain investment tendencies, thereby determining the target user's risk assessment information.

[0057] According to an embodiment of the present disclosure, the financial product information may include the risk level, fund manager, asset size, purchase cycle and redemption cycle of the financial product, etc.

[0058] According to the embodiments of the present disclosure, the acquired financial behavior information of the target user may include sensitive information involving the privacy or confidentiality of the target user, such as transaction amount information in the user's flow information, etc. Since directly using sensitive information for financial information recommendation or training of information recommendation models may bring information security risks, the embodiments of the present disclosure adopt a data desensitization method in the process of acquiring financial behavior information.

[0059] According to an embodiment of the present disclosure, obtaining target financial behavior information of a target user may include the following operations.

[0060] Obtain the original financial behavior information of the target user; determine the user sensitive information related to the user privacy in the original financial behavior information; process the user sensitive information using a data desensitization method to obtain processed user sensitive information; and determine the target financial behavior information based on the original financial behavior information and the processed user sensitive information.

[0061] For example, for the transaction amount in the original financial behavior information, the transaction amount can be logarithmized and rounded, as shown in formula (1):

[0062] Y′=[log Y] (1)

[0063] In the formula, Y represents the original transaction amount, Y′ represents the de-sensitized transaction amount, and [·] represents rounding down.

[0064] It should be noted that the desensitization method shown in formula (1) is only an exemplary embodiment. On the basis of ensuring that the desensitized transaction amount can still maintain a partial order relationship consistent with the original transaction amount, any mathematical method can be used to desensitize the data. For example, before performing a logarithmic operation on the original transaction amount, a linear transformation can also be performed on the original transaction amount.

[0065] According to the embodiments of the present disclosure, by using the data desensitization method, on the one hand, the transaction amount after desensitization cannot be directly and accurately restored to the original transaction amount, which effectively protects the user's information security and effectively avoids the information leakage problem that may occur during the use of the information recommendation model. On the other hand, since the distribution between the original transaction amount and the transaction volume is usually a power exponential distribution, it is not conducive to model learning. Therefore, by using the data desensitization method on the original transaction amount, the distribution between the desensitized transaction amount and the transaction volume can also be made close to the normal distribution, which is conducive to improving the accuracy of the information recommendation model.

[0066] Figure 3 The following is a schematic diagram schematically showing the information recommendation model training process according to an embodiment of the present disclosure.

[0067] like Figure 3 As shown, the information recommendation model can be obtained by writing model training code based on open source frameworks such as tensorflow, and using the desensitized target financial behavior information 310 and label information 320 of the sample user to run the training task on a training cluster where the open source framework is deployed. The training cluster can be deployed on a private cloud or a confidential computer to avoid leakage of user information.

[0068] According to an embodiment of the present disclosure, the target financial behavior information 310 of the sample user after desensitization includes the user flow information 311, user risk assessment information 312 and financial product information 313 of the sample user; the label information 320 includes the real click rate 321, the real display information 322 and the real benefit 323.

[0069] In addition, the information recommendation model of the embodiment of the present disclosure can also be jointly modeled with other institutions through the framework of federated learning. Through the federated transfer learning method, it is possible to make full use of the user's scattered data in various institutions without leaking privacy, thereby providing users with more accurate information recommendation services.

[0070] According to an embodiment of the present disclosure, the information recommendation model includes a click rate model 330 and a revenue model 340 .

[0071] According to an embodiment of the present disclosure, the click rate model 330 may be obtained by jointly training a first long short-term memory neural network model 331 , a first deep neural network model 332 , and a first neural network model 333 .

[0072] According to an embodiment of the present disclosure, the first training sample used in training the click rate model 330 may include multiple positive samples and multiple negative samples. Each positive sample may include target financial behavior information 310, and use the real click rate 321 of the sample user for the sample product as a label; each negative sample may include target financial behavior information 310, and use the real display information 322 of the sample user for the sample product as a label.

[0073] According to an embodiment of the present disclosure, each iteration process of training the click rate model 330 is as follows:

[0074] First, multiple positive samples in the first training sample are input into the click rate model 330 to obtain the predicted click rate.

[0075] According to an embodiment of the present disclosure, the user flow information 311 of the sample user included in each positive sample among multiple positive samples can be input into the first long short-term memory neural network model 331, and the user risk assessment information 312 and financial product information 313 of the sample user included in each positive sample among multiple positive samples can be input into the first deep neural network model 332.

[0076] According to an embodiment of the present disclosure, a third vector may be obtained by concatenating a first vector output by the first long short-term memory neural network model 331 and a second vector output by the first deep neural network model 332.

[0077] According to an embodiment of the present disclosure, the third vector is input into the first neural network model 333, and the predicted click rate of the sample user for the sample product included in the output positive sample can be obtained.

[0078] Then, multiple negative samples in the first training sample are input into the click rate model 330 to obtain predicted display information.

[0079] According to an embodiment of the present disclosure, the user flow information 311 of the sample user included in each of the multiple negative samples can be input into the first long short-term memory neural network model 331, and the user risk assessment information 312 and financial product information 313 of the sample user included in each of the multiple negative samples can be input into the first deep neural network model 332.

[0080] According to an embodiment of the present disclosure, a sixth vector may be obtained by concatenating the fourth vector output by the first long short-term memory neural network model 331 and the fifth vector output by the first deep neural network model 332.

[0081] According to an embodiment of the present disclosure, the sixth vector is input into the first neural network model 333, and the predicted display information of the sample user for the sample product included in the output negative sample can be obtained.

[0082] Finally, the loss function is used to complete the iteration of the model parameters of the click rate model 330.

[0083] According to an embodiment of the present disclosure, a first loss function 334 can be used to sequentially process multiple real click rates 321 and multiple predicted click rates, and multiple real display information 322 and multiple predicted display information to obtain loss values ​​for completing the iteration of model parameters.

[0084] According to an embodiment of the present disclosure, the positive samples and negative samples may be alternately input into the first long short-term memory neural network model, the first deep neural network model and the first neural network model to train the above network models to obtain the click rate model 330.

[0085] According to the embodiments of the present disclosure, since positive samples and negative samples are input into the first long short-term memory neural network model, the first deep neural network model and the first neural network model to train the above network models, the above network models can learn the characteristics of positive samples and negative samples by comparison. Since the characteristics of positive samples and negative samples can be learned by comparison, the effective recognition of positive samples is achieved, thereby ensuring the prediction accuracy of the click-through rate model. In addition, the positive samples and negative samples are alternately input into the above network model to train the above network model, which also speeds up the model training speed.

[0086] According to an embodiment of the present disclosure, the profit model 340 may be obtained by joint training of a second long short-term memory neural network model 341 , a second deep neural network model 342 , and a second neural network model 343 .

[0087] According to an embodiment of the present disclosure, the second training sample used in training the benefit model 340 includes multiple training groups, wherein each training group includes target financial behavior information 310, and uses the real benefits 323 of the sample user for the sample product as a label.

[0088] According to an embodiment of the present disclosure, each iteration process of the training revenue model 340 is as follows:

[0089] First, multiple training groups in the second training sample are input into the profit model 340 to obtain predicted benefits.

[0090] According to an embodiment of the present disclosure, the user flow information 311 of the sample users included in each training group in the second training sample may be input into the second long short-term memory neural network model 341 to obtain the output seventh vector.

[0091] According to an embodiment of the present disclosure, the user risk assessment information 312 and the financial product information 313 of the sample users included in each training group in the second training sample may be input into the first deep neural network model 342 to obtain an output eighth vector.

[0092] According to an embodiment of the present disclosure, the ninth vector obtained by concatenating the seventh vector and the eighth vector is input into the second neural network model 343, and the predicted benefits of the sample users included in the output training group for the sample products can be obtained.

[0093] Afterwards, the loss function is used to complete the iteration of the model parameters of the profit model 340.

[0094] According to an embodiment of the present disclosure, the second loss function 344 may be used to process multiple real benefits 323 and multiple predicted benefits to obtain loss values ​​for completing the iteration of model parameters.

[0095] According to an embodiment of the present disclosure, the first loss function 334 and the second loss function 344 may be, for example, a log loss function, a square loss function, a hubor loss function, etc. The first loss function 334 and the second loss function 344 may use the same loss function or different loss functions.

[0096] According to an embodiment of the present disclosure, during the training process of the click rate model 330 and the revenue model 340, methods such as the gradient descent method and the stochastic gradient descent method may be used to change the model parameters.

[0097] According to an embodiment of the present disclosure, the end condition of the model training can be preset in the process of training the click rate model 330 and the revenue model 340. For example, the training can be stopped when the loss value of the loss function in each iteration is less than a preset value, or the training can be stopped when the gradient of the loss function in each iteration is less than a preset value, or the training can be stopped after the number of training times reaches a preset value.

[0098] According to the embodiment of the present disclosure, after the training of the click rate model 330 and the benefit model 340 is completed, the target financial behavior information 310 of the target user can be input into the click rate model 330 and the benefit model 340 respectively, and the click rate 351 and benefit 352 for each candidate financial information in the plurality of candidate financial information are output. According to the click rate 351 and benefit 352 of each candidate financial information, the recommendation score 353 of the candidate financial information can be determined. For example, the recommendation score can be calculated according to the method of formula (2):

[0099] score=ctr×value (2)

[0100] In the formula, score represents the recommendation score of the candidate financial information, ctr represents the click rate of the candidate financial information, and value represents the benefit of the candidate financial information.

[0101] According to an embodiment of the present disclosure, the click rate of candidate financial information indicates the probability of a user using the candidate financial information after receiving it, and the benefit of the candidate financial information indicates the income of the information provider based on the recommendation of the financial information. Depending on the specific form of the financial information, the click rate and benefit of the financial information are also different. For example, if the candidate financial information indicates a specific financial product, the click rate of the candidate financial information indicates the probability of the user selecting the financial product after receiving the recommendation of the financial product; the benefit of the candidate financial information indicates the amount of money the user deposits after receiving the recommendation of the financial product.

[0102] According to an embodiment of the present disclosure, the target financial information can be determined based on the recommendation scores 353 of multiple candidate financial information, and the target financial information can be recommended to the target user. For example, the multiple recommendation scores can be sorted, and the candidate financial information with the highest score can be selected from the sorted results as the target financial information. In addition, a classifier can also be used to determine the target financial information, for example, a softmax classifier can be used to determine the target financial information.

[0103] According to an embodiment of the present disclosure, a financial information recommendation method is provided based on deep learning to process financial behavior information. Since the financial behavior information includes the user flow information, user risk assessment information and financial product information of the target user, the above information can reflect the changes in the user's risk-bearing capacity, cash flow status and investment and financial management habits. Therefore, the method can automatically learn the changes in the user's risk-bearing capacity, cash flow status and investment and financial management habits. As a result, it can provide users with more personalized, higher-level and more suitable financial information recommendation services for their current situations, thereby reducing the cost for different users to obtain suitable financial information, thereby meeting the needs of users and improving the user experience.

[0104] Figure 4 The block diagram of the information recommendation device 400 according to an embodiment of the present disclosure is schematically shown.

[0105] like Figure 4 As shown, the information recommendation device 400 includes an acquisition module 410 , an output module 420 , a determination module 430 and a recommendation module 440 .

[0106] The acquisition module 410 is used to acquire the target financial behavior information of the target user, wherein the target financial behavior information of the target user includes at least one of the following: user flow information, user risk assessment information and financial product information of the target user.

[0107] The output module 420 is used to input the target financial behavior information of the target user into the information recommendation model, and output a recommendation score for each candidate financial information among the multiple candidate financial information.

[0108] The determination module 430 is used to determine target financial information from multiple candidate financial information according to multiple recommendation scores.

[0109] The recommendation module 440 is used to recommend target financial information to target users.

[0110] According to the embodiments of the present disclosure, the target financial behavior of the target user is processed by using the information recommendation model to obtain the recommendation score of each candidate financial information in multiple candidate financial information, and the target financial information recommended to the target user is determined based on the multiple recommendation scores. Since the target financial behavior information of the target user can reflect the personalized financial information of the user, the target financial behavior information of the target user is processed by using the information recommendation model to obtain the recommendation score of each candidate financial information in multiple candidate financial information, and the target financial information recommended to the target user is determined based on the multiple recommendation scores, thereby realizing personalized recommendation, and the target financial behavior information is processed by using the information recommendation model to improve the accuracy of the recommendation, thereby meeting the needs of the user.

[0111] According to an embodiment of the present disclosure, the acquisition module 410 includes a first acquisition unit, a second acquisition unit, a third acquisition unit and a fourth acquisition unit.

[0112] A first acquisition unit, used to acquire original financial behavior information of the user;

[0113] A second acquisition unit, used to determine user sensitive information related to user privacy in the original financial behavior information;

[0114] A third acquisition unit is used to process the user sensitive information by using a data desensitization method to obtain the processed user sensitive information; and

[0115] The fourth acquisition unit is used to determine the target financial behavior information according to the original financial behavior information and the processed user sensitive information.

[0116] According to an embodiment of the present disclosure, the information recommendation model includes a click-through rate model and a revenue model, and the information recommendation device also includes a first training module and a second training module, wherein:

[0117] A first training module is used to obtain a click rate model by jointly training the first long short-term memory neural network model, the first deep neural network model and the first neural network model;

[0118] The first training module is used to obtain a profit model by jointly training the second long short-term memory neural network model, the second deep neural network model and the second neural network model.

[0119] According to an embodiment of the present disclosure, the output module 420 includes a first output unit, a second output unit and a third output unit.

[0120] A first output unit, configured to input the target financial behavior information of the target user into a click rate model, and output a click rate for each candidate financial information among a plurality of candidate financial information;

[0121] A second output unit, configured to input the target financial behavior information of the target user into the benefit model, and output the benefit for each candidate financial information among the plurality of candidate financial information; and

[0122] The third output unit is used to determine a recommendation score for the candidate financial information according to the click rate and benefit of each candidate financial information among the plurality of candidate financial information.

[0123] According to an embodiment of the present disclosure, the first training module includes a first training unit, a second training unit, a third training unit, a fourth training unit, a fifth training unit, a sixth training unit, a seventh training unit, an eighth training unit, a ninth training unit, and a tenth training unit.

[0124] in:

[0125] A first training unit is used to obtain a first training sample, wherein the first training sample includes a plurality of positive samples and a plurality of negative samples, each positive sample includes target financial behavior information of a sample user and a real click rate of a sample user for a sample product, each negative sample includes target training financial behavior information of a sample user and real display information of a sample user for a sample product, and the target training financial behavior information of a sample user includes user flow information, user risk assessment information, and financial product information of the sample user;

[0126] A second training unit is used to input user flow information of a sample user included in each positive sample of the plurality of positive samples into a first long short-term memory neural network model, and output a first vector;

[0127] A third training unit is used to input the user risk assessment information and financial product information of the sample user included in each positive sample of the multiple positive samples into the first deep neural network model, and output a second vector;

[0128] A fourth training unit is used to concatenate the second vector and the second vector to obtain a third vector;

[0129] a fifth training unit, configured to input the third vector into the first neural network model, and output a predicted click rate of the sample user included in the positive sample for the sample product;

[0130] a sixth training unit, inputting user flow information of a sample user included in each negative sample of the plurality of negative samples into the first long short-term memory neural network model, and outputting a fourth vector;

[0131] a seventh training unit, inputting the user risk assessment information and the financial product information of the sample user included in each negative sample of the plurality of negative samples into the first deep neural network model, and outputting a fifth vector;

[0132] The eighth training unit concatenates the fourth vector and the fifth vector to obtain a sixth vector;

[0133] a ninth training unit, inputting the sixth vector into the first neural network model, and outputting predicted display information of the sample user included in the negative sample for the sample product; and

[0134] The tenth training unit trains the first long short-term memory neural network model, the first deep neural network model and the first neural network model according to multiple real click-through rates, multiple predicted click-through rates, multiple real display information and multiple predicted display information to obtain a click-through rate model.

[0135] According to an embodiment of the present disclosure, the second training module includes an eleventh training unit, a twelfth training unit, a thirteenth training unit, a fourteenth training unit, a fifteenth training unit and a sixteenth training unit.

[0136] an eleventh training unit, configured to obtain a second training sample, wherein the second training sample includes a plurality of training groups, each training group includes target financial behavior information of a sample user and actual benefits of the sample user for a sample product, and the target training financial behavior information of the sample user includes user flow information, user risk assessment information, and financial product information of the sample user;

[0137] a twelfth training unit, configured to input user flow information of sample users included in each of the plurality of training groups into a second long short-term memory neural network model, and output a seventh vector;

[0138] A thirteenth training unit, configured to input the user risk assessment information and financial product information of the sample users included in each training group in the plurality of training groups into the second deep neural network model, and output an eighth vector;

[0139] A fourteenth training unit is used to concatenate the seventh vector and the eighth vector to obtain a ninth vector;

[0140] a fifteenth training unit, configured to input the ninth vector into the second neural network model, and output predicted benefits of the sample users included in the training group for the sample products; and

[0141] The sixteenth training unit is used to train the second long short-term memory neural network model, the second deep neural network model and the second neural network model according to multiple real benefits and multiple predicted benefits to obtain a profit model.

[0142] According to an embodiment of the present disclosure, the determination module 430 includes a first determination unit and a second determination unit.

[0143] A first determining unit is used to sort the multiple recommendation scores to obtain a sorting result; and

[0144] The second determining unit is used to determine target financial information from multiple candidate financial information according to the sorting result.

[0145] According to an embodiment of the present disclosure, user transaction information includes at least one of the following: transaction time, transaction amount and transaction nature of transactions within a preset time period; financial product information includes at least one of the following: risk level of the financial product, fund manager, asset size, buying cycle and redemption cycle.

[0146] According to an embodiment of the present disclosure, the first training module and the second training module utilize a federated learning framework when training the information recommendation model.

[0147] According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits, or at least part of the functions of any one of them can be implemented in one module. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be split into multiple modules for implementation. According to the embodiments of the present invention, any one or more of the modules, submodules, units, and subunits can be at least partially implemented as hardware circuits, such as field programmable gate arrays (Field Programmable Gate Array, FPGA), programmable logic arrays (Programmable Logic Arrays, PLA), systems on chips, systems on substrates, systems on packages, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), or can be implemented by hardware or firmware of any other reasonable way of integrating or packaging the circuit, or by any one of the three implementation methods of software, hardware, and firmware or by a suitable combination of any of them. Alternatively, according to the embodiments of the present invention, one or more of the modules, submodules, units, and subunits can be at least partially implemented as computer program modules, and when the computer program modules are run, the corresponding functions can be performed.

[0148] For example, any multiple of the acquisition module 410, the output module 420, the determination module 430, and the recommendation module 440 can be combined in one module / unit / sub-unit for implementation, or any one of the modules / units / sub-units can be split into multiple modules / units / sub-units. Alternatively, at least part of the functions of one or more of these modules / units / sub-units can be combined with at least part of the functions of other modules / units / sub-units and implemented in one module / unit / sub-unit. According to an embodiment of the present disclosure, at least one of the acquisition module 410, the output module 420, the determination module 430, and the recommendation module 440 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware or in a suitable combination of any of them. Alternatively, at least one of the acquisition module 410 , the output module 420 , the determination module 430 , and the recommendation module 440 may be at least partially implemented as a computer program module, and when the computer program module is executed, a corresponding function may be performed.

[0149] According to the embodiments of the present disclosure, the information recommendation device part in the embodiments of the present disclosure corresponds to the information recommendation method part in the embodiments of the present disclosure. The description of the information recommendation device part specifically refers to the information recommendation method part, which will not be repeated here.

[0150] Figure 5 A block diagram of an electronic device 500 suitable for implementing the method described above according to an embodiment of the present disclosure is schematically shown. Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0151] like Figure 5 As shown, the electronic device 500 according to the embodiment of the present disclosure includes a processor 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 to the random access memory (RAM) 503. The processor 501 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 501 may also include an onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to the embodiment of the present disclosure.

[0152] In RAM 503, various programs and data required for the operation of electronic device 500 are stored. Processor 501, ROM 502 and RAM 503 are connected to each other via bus 504. Processor 501 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 502 and / or RAM 503. It should be noted that the program can also be stored in one or more memories other than ROM 502 and RAM 503. Processor 501 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0153] According to an embodiment of the present disclosure, the electronic device 500 may further include an input / output (I / O) interface 505, which is also connected to the bus 504. The electronic device 500 may further include one or more of the following components connected to the I / O interface 505: an input portion 506 including a keyboard, a mouse, etc.; an output portion 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 508 including a hard disk, etc.; and a communication portion 509 including a network interface card such as a LAN card, a modem, etc. The communication portion 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage portion 508 as needed.

[0154] According to an embodiment of the present disclosure, the method flow according to an embodiment of the present disclosure can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the processor 501, the above-mentioned functions defined in the system of the embodiment of the present disclosure are executed. According to an embodiment of the present disclosure, the system, equipment, device, module, unit, etc. described above can be implemented by a computer program module.

[0155] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0156] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium. For example, it may include, but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus, or a device.

[0157] For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 502 and / or the RAM 503 described above and / or one or more memories other than the ROM 502 and the RAM 503 .

[0158] An embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains a program code for executing the method provided by the embodiment of the present disclosure. When the computer program product runs on an electronic device, the program code is used to enable the electronic device to implement the information recommendation method provided by the embodiment of the present disclosure.

[0159] When the computer program is executed by the processor 501, the above functions defined in the system / device of the embodiment of the present disclosure are executed. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0160] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 509, and / or installed from the removable medium 511. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0161] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the account computing device, partially on the account device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the account computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0162] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. It can be understood by those skilled in the art that the features recorded in the various embodiments and / or claims of the present disclosure can be combined and / or combined in a variety of ways, even if such a combination or combination is not explicitly recorded in the present disclosure. In particular, without departing from the spirit and teaching of the present disclosure, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0163] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. An information recommendation method, comprising: Acquire target financial behavior information of a target user, wherein the target financial behavior information of the target user includes: user flow information, user risk assessment information and financial product information of the target user; Inputting the target financial behavior information of the target user into an information recommendation model, and outputting a recommendation score for each candidate financial information among a plurality of candidate financial information; determining target financial information from the plurality of candidate financial information according to the plurality of recommendation scores; and recommending the target financial information to the target user; Wherein, the information recommendation model includes a click-through rate model and a revenue model; The step of inputting the target financial behavior information of the target user into an information recommendation model and outputting a recommendation score for each of the plurality of candidate financial information includes: Inputting the target financial behavior information of the target user into the click rate model, and outputting the click rate for each candidate financial information among the plurality of candidate financial information; Inputting the target financial behavior information of the target user into the benefit model, and outputting the benefit for each candidate financial information among the plurality of candidate financial information; and determining a recommendation score for each candidate financial information according to a click rate and a benefit of each candidate financial information among the plurality of candidate financial information; The revenue model is trained by the following methods, including: Acquire a second training sample, wherein the second training sample includes a plurality of training groups, each of the training groups includes target financial behavior information of the sample user and the real benefits of the sample user for a sample product, and the target training financial behavior information of the sample user includes user flow information, user risk assessment information, and financial product information of the sample user; Inputting user flow information of sample users included in each of the plurality of training groups into a second long short-term memory neural network model, and outputting a seventh vector; Inputting the user risk assessment information and financial product information of the sample users included in each training group in the plurality of training groups into a second deep neural network model, and outputting an eighth vector; Concatenate the seventh vector and the eighth vector to obtain a ninth vector; Inputting the ninth vector into a second neural network model, and outputting predicted benefits of the sample users included in the training group for the sample products; and The second long short-term memory neural network model, the second deep neural network model and the second neural network model are trained according to the multiple real benefits and the multiple predicted benefits to obtain the profit model.

2. The method according to claim 1, wherein: The obtaining of target financial behavior information of the target user includes: Obtaining original financial behavior information of the target user; Determining user sensitive information related to user privacy in the original financial behavior information; Processing the user sensitive information using a data desensitization method to obtain processed user sensitive information; and The target financial behavior information is determined according to the original financial behavior information and the processed user sensitive information.

3. The method according to claim 1, wherein: The click rate model is obtained by jointly training a first long short-term memory neural network model, a first deep neural network model and a first neural network model.

4. The method according to claim 3, wherein: The click rate model is obtained by jointly training a first long short-term memory neural network model, a first deep neural network model and a first neural network model, and includes: Acquire a first training sample, wherein the first training sample includes a plurality of positive samples and a plurality of negative samples, each of the positive samples includes target financial behavior information of a sample user and a real click rate of the sample user for a sample product, each of the negative samples includes target training financial behavior information of the sample user and real display information of the sample user for the sample product, and the target training financial behavior information of the sample user includes user flow information, user risk assessment information, and financial product information of the sample user; Inputting user flow information of a sample user included in each positive sample of the plurality of positive samples into the first long short-term memory neural network model, and outputting a first vector; Inputting the user risk assessment information and financial product information of the sample user included in each positive sample of the plurality of positive samples into the first deep neural network model, and outputting a second vector; concatenate the second vector and the second vector to obtain a third vector; Inputting the third vector into the first neural network model, and outputting the predicted click rate of the sample user included in the positive sample for the sample product; Inputting user flow information of a sample user included in each negative sample of the multiple negative samples into the first long short-term memory neural network model, and outputting a fourth vector; Inputting the user risk assessment information and financial product information of the sample user included in each negative sample of the multiple negative samples into the first deep neural network model, and outputting a fifth vector; Concatenate the fourth vector and the fifth vector to obtain a sixth vector; Inputting the sixth vector into the first neural network model, and outputting the predicted display information of the sample user included in the negative sample for the sample product; and The first long short-term memory neural network model, the first deep neural network model and the first neural network model are trained according to the multiple real click-through rates, the multiple predicted click-through rates, the multiple real display information and the multiple predicted display information to obtain the click-through rate model.

5. The method according to claim 1, wherein: The step of determining target financial information from the plurality of candidate financial information according to the plurality of recommendation scores includes: Sorting the plurality of recommendation scores to obtain a sorting result; and The target financial information is determined from the plurality of candidate financial information according to the ranking result.

6. The method according to any one of claims 1 to 5, wherein The user transaction information includes at least one of the following: transaction time, transaction amount and transaction nature of transactions within a preset time period; The financial product information includes at least one of the following: the risk level of the financial product, the fund manager, the asset size, the purchase cycle and the redemption cycle.

7. The method according to claim 1, wherein: The information recommendation model is trained using a federated learning framework.

8. An information recommendation device, comprising: An acquisition module, used to acquire target financial behavior information of a target user, wherein the target financial behavior information of the target user includes: user flow information, user risk assessment information and financial product information of the target user; An output module, configured to input the target financial behavior information of the target user into an information recommendation model, and output a recommendation score for each candidate financial information among a plurality of candidate financial information; a determination module, configured to determine target financial information from the plurality of candidate financial information according to the plurality of recommendation scores; and A recommendation module, used for recommending the target financial information to the target user; Wherein, the information recommendation model includes a click rate model and a revenue model, and the information recommendation device also includes a first training module and a second training module; The output module comprises: A first output unit, configured to input the target financial behavior information of the target user into the click rate model, and output a click rate for each candidate financial information among the plurality of candidate financial information; a second output unit, configured to input the target financial behavior information of the target user into the benefit model, and output the benefit for each candidate financial information among the plurality of candidate financial information; and a third output unit, configured to determine a recommendation score for each candidate financial information according to the click rate and benefit of each candidate financial information among the plurality of candidate financial information; The second training module includes: an eleventh training unit, configured to obtain a second training sample, wherein the second training sample includes a plurality of training groups, each of the training groups includes target financial behavior information of the sample user and actual benefits of the sample user for a sample product, and the target training financial behavior information of the sample user includes user flow information, user risk assessment information, and financial product information of the sample user; A twelfth training unit, configured to input user flow information of sample users included in each of the plurality of training groups into a second long short-term memory neural network model, and output a seventh vector; A thirteenth training unit, configured to input the user risk assessment information and financial product information of the sample users included in each of the plurality of training groups into a second deep neural network model, and output an eighth vector; A fourteenth training unit, configured to concatenate the seventh vector and the eighth vector to obtain a ninth vector; a fifteenth training unit, configured to input the ninth vector into a second neural network model, and output predicted benefits of the sample users included in the training group for the sample products; and The sixteenth training unit is used to train the second long short-term memory neural network model, the second deep neural network model and the second neural network model according to the multiple real benefits and the multiple predicted benefits to obtain the profit model.

9. An electronic device, comprising: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enables the processor to implement the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, wherein the computer program is used to implement the method according to any one of claims 1 to 7 when executed by a processor.

Images