Fund recommendation method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a fund recommendation method and device, equipment and a storage medium, and the method comprises the steps: obtaining portrait associated information of a target object; inputting the portrait associated information into a pre-trained target fund recommendation model to obtain a target recommendation result of the target object; wherein the target fund recommendation model comprises at least one target fund product and a target purchase amount interval corresponding to the target fund product. According to the technical scheme, the problems that in the prior art, only fund products are recommended, and the recommendation content is single are solved, the fund products and purchase amount requirements of the target object can be predicted, and therefore services and promotion are provided in a targeted mode, the user experience is improved, and the promotion effect is improved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of financial recommendation, and in particular to a fund recommendation method, device, equipment and storage medium. Background Art

[0002] As an important financial product, fund products occupy a key position among various financial products sold by banks, and are also an important bridge for banks to increase customer stickiness and generate revenue. In recent years, with the popularization of financial knowledge, consumers' acceptance of fund-type financial products has increased year by year, so the promotion of fund products has become particularly important.

[0003] In the field of fund promotion, the existing fund purchase intention prediction model only predicts whether customers will purchase certain funds. It can only push reminders for the fund and the recommendation content is relatively simple, which cannot meet the increasingly diverse fund recommendation needs. Summary of the invention

[0004] The embodiments of the present invention provide a fund recommendation method, apparatus, device and storage medium, which can predict the fund products and purchase amount requirements of the target object, thereby providing targeted services and promotions, improving user experience and enhancing promotion effects.

[0005] In a first aspect, an embodiment of the present invention provides a fund recommendation method, the method comprising:

[0006] Obtain portrait association information of the target object; input the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product.

[0007] In a second aspect, an embodiment of the present invention provides a fund recommendation device, the device comprising:

[0008] Obtain portrait association information of the target object; input the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product.

[0009] In a third aspect, an embodiment of the present invention provides a computer device, the computer device comprising:

[0010] one or more processors;

[0011] A memory for storing one or more programs;

[0012] When the one or more programs are executed by the one or more processors, the one or more processors implement the fund recommendation method described in any embodiment.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the fund recommendation method described in any embodiment.

[0014] The technical solution provided by the embodiment of the present invention obtains the portrait association information of the target object; inputs the portrait association information into a pre-trained target fund recommendation model to obtain the target recommendation result of the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product. The technical solution of the embodiment of the present invention solves the problem that the prior art only recommends fund products and the recommendation content is relatively single, and can predict the fund product and purchase amount needs of the target object, thereby providing targeted services and promotions, improving user experience, and improving promotion effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a fund recommendation method provided by an embodiment of the present invention;

[0016] Figure 2 This is another flow chart of a fund recommendation method provided by an embodiment of the present invention;

[0017] Figure 3 This is a workflow diagram for fund recommendation provided by an embodiment of the present invention;

[0018] Figure 4 It is a structural schematic diagram of a fund recommendation device provided by an embodiment of the present invention;

[0019] Figure 5 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Figure 1It is a flow chart of a fund recommendation method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of recommending fund products to target objects. The method can be executed by a fund recommendation device, which can be implemented by software and / or hardware.

[0022] like Figure 1 As shown, the fund recommendation method includes the following steps:

[0023] S110, obtaining portrait-related information of the target object.

[0024] Among them, the target object may be an object that needs to be recommended for a fund. The portrait-related information may be information that has an association relationship with the portrait of the target object. Exemplarily, the portrait-related information may include basic information of the object (such as age, gender, occupation, etc.), object behavior information (such as product preference, channel preference, transaction history, abnormal operation, risky transaction, etc.), object value information (such as whether it is a high-net-worth object, marketable potential, financial assets, etc.) and marketing feedback information (historical marketing records, push information click records, recommended page browsing time, outbound marketing call duration), etc. By obtaining the portrait-related information of the target object, it is convenient to determine the fund products that are suitable for the target object and the purchase amount range of the fund products based on the portrait-related information.

[0025] S120: Input the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object.

[0026] Among them, the target recommendation result can be a fund recommendation result adapted for the target object. Specifically, the target fund product recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product. Among them, the target fund product can be a fund product recommended to the target object for purchase. The target purchase amount range can be a purchase amount range for the target fund product recommended to the target object. Specifically, the portrait association information can be input into a pre-trained target fund recommendation model so that the model analyzes the portrait association information and then obtains the target recommendation result corresponding to the target object.

[0027] The technical solution provided by the embodiment of the present invention obtains the portrait association information of the target object; inputs the portrait association information into a pre-trained target fund recommendation model to obtain the target recommendation result of the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product. The technical solution of the embodiment of the present invention solves the problem that the prior art only recommends fund products and the recommendation content is relatively single, and can predict the fund product and purchase amount needs of the target object, thereby providing targeted services and promotions, improving user experience, and improving promotion effects.

[0028] Figure 2 It is another fund recommendation method flow chart provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of recommending fund products to the target object. Based on the above embodiment, this embodiment further illustrates how to train a target fund recommendation model, and recommend funds to the target object based on the trained target fund recommendation model. The device can be implemented by software and / or hardware and integrated in a computer device with application development function.

[0029] like Figure 2 As shown, the fund recommendation method includes the following steps:

[0030] S210. Obtain preliminary fund recommendation model and preset portrait sample information.

[0031] The preliminary fund recommendation model may be an original fund recommendation model that has not been pre-trained. Furthermore, the preset portrait sample information may be preset sample information about the portrait of the target object. The preset portrait sample information may contain corresponding recommended fund labels and purchase amount range labels, which are used to evaluate the recommendation results of the initial fund recommendation model and adjust the model according to the evaluation results.

[0032] S220. Training the preliminary fund recommendation model based on the preset portrait sample information to obtain the target fund recommendation model.

[0033] The target fund recommendation model may be a model that is ultimately used to recommend funds to the target object. The target fund recommendation model may be obtained by training a preliminary fund recommendation model based on preset portrait sample information.

[0034] Optionally, the preliminary fund recommendation model is trained based on the preset portrait sample information to obtain the target fund recommendation model, including: inputting the preset portrait sample information into the preliminary fund recommendation model to obtain a sample recommendation result; determining a recommendation loss function based on the sample recommendation result and a preset recommendation label of the preset portrait sample information; and adjusting the preliminary fund recommendation model according to the recommendation loss function to obtain the target fund recommendation model.

[0035] The sample recommendation result may be the fund recommendation result of the preliminary fund recommendation model for the preset portrait sample information. The preset recommended fund label and the purchase amount interval label may be labels pre-marked in the preset portrait sample information. By comparing the sample recommendation result with the preset recommended fund label and the purchase amount interval label of the preset portrait sample information, the error of the preliminary fund recommendation model in making fund recommendations may be known.

[0036] Furthermore, a recommendation loss function can be determined based on the comparison results. The recommendation loss function can be a loss function corresponding to the recommendation error. The error size can be known through the recommendation loss function. Subsequently, the preliminary fund recommendation model can be adjusted according to the recommendation loss function to obtain the target fund recommendation model. Specifically, the model parameters in the preliminary fund recommendation model can be adjusted according to the recommendation loss function to obtain the target fund recommendation model.

[0037] S230: Obtain image-related information of the target object.

[0038] Among them, the target object may be an object that needs to be recommended for a fund. The portrait-related information may be information that has an association relationship with the portrait of the target object. Exemplarily, the portrait-related information may include basic information of the object (such as age, gender, occupation, etc.), object behavior information (such as product preference, channel preference, transaction history, abnormal operation, risky transaction, etc.), object value information (such as whether it is a high-net-worth object, marketable potential, financial assets, etc.) and marketing feedback information (historical marketing records, push information click records, recommended page browsing time, outbound marketing call duration), etc. By obtaining the portrait-related information of the target object, it is convenient to determine the fund products that are suitable for the target object and the purchase amount range of the fund products based on the portrait-related information.

[0039] S240: Input the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object.

[0040] Among them, the target recommendation result can be a fund recommendation result adapted for the target object. Specifically, the target fund product recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product. Among them, the target fund product can be a fund product recommended to the target object for purchase. The target purchase amount range can be a purchase amount range for the target fund product recommended to the target object. Specifically, the portrait association information can be input into a pre-trained target fund recommendation model so that the model analyzes the portrait association information and then obtains the target recommendation result corresponding to the target object.

[0041] Optionally, the target fund recommendation model includes a portrait feature extraction network and a portrait feature analysis network, and the portrait association information is input into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object, including: inputting the portrait association information into the portrait feature extraction network to obtain a target portrait feature of the target object; inputting the target portrait feature into the portrait feature analysis network to obtain the target recommendation result.

[0042] Among them, the portrait feature extraction network can be used to extract the features of the input portrait information. The target portrait feature can be feature information about the portrait-related information. Among them, the target portrait feature includes: at least one of the basic features of the object, the behavior features of the object, the value features of the object, and the recommendation feedback features. Specifically, the portrait-related information can be input into the portrait feature extraction network so that the network can extract features from the portrait-related information to obtain the target portrait features. Furthermore, the portrait feature analysis network can analyze the target portrait features, determine the fund products and purchase amount ranges that match the target portrait features, and obtain the target recommendation results.

[0043] Optionally, the portrait feature analysis network includes: a fund recommendation sub-network and an amount recommendation sub-network, and the target portrait feature is input into the portrait feature analysis network to obtain the target recommendation result, including: inputting the target portrait feature into the fund recommendation sub-network to obtain at least one target fund product; inputting the target fund product and the target portrait feature into the amount recommendation sub-network to obtain the target purchase amount range corresponding to the target fund product.

[0044] Among them, the fund recommendation subnetwork can be used to determine the fund products that are compatible with the portrait features. Specifically, the target portrait features can be input into the fund recommendation subnetwork so that the network matches the fund products corresponding to the target portrait features, thereby obtaining the target fund products. The amount recommendation subnetwork can be used to determine the purchase amount range of the target fund product. Specifically, the identification information and the target portrait features of the target fund product can be input into the amount recommendation subnetwork so that the network can determine the target purchase amount range corresponding to each target fund product based on the target portrait features.

[0045] Optionally, the preliminary fund recommendation model also includes a marketing recommendation network. After determining the target recommendation result, the target recommendation result can also be input into the marketing recommendation network to obtain the marketing recommendation result corresponding to the target recommendation result.

[0046] The sales promotion prediction network can be used to determine the sales promotion type corresponding to the target recommendation result. The sales promotion recommendation result can be a recommendation result of a sales promotion strategy adapted to the target fund product. Specifically, after the target recommendation result is determined, the target recommendation result can be input into the sales promotion recommendation network so that the network matches the sales promotion strategy type corresponding to the target recommendation result, and the obtained sales promotion strategy type is used as the sales promotion prediction result. The sales promotion strategy type can be a red envelope marketing such as a 10 yuan discount for a purchase of 1,000 yuan.

[0047] The technical solution of the embodiment of the present invention has the following beneficial effects:

[0048] (1) Based on the existing solutions, the present invention introduces machine learning technology to predict the purchase amount range of the target fund, broadens the prediction range, realizes precise marketing in the amount dimension, and broadens the marketing means.

[0049] (2) The present invention introduces a marketing effect feedback mechanism, so as to provide marketing effect feedback based on prediction, incorporate marketing effect information into prediction, form a data closed loop, improve data utilization, and improve prediction accuracy.

[0050] For example, in order to better understand the technical solution provided by the present invention, a specific embodiment is introduced below: Figure 3 FIG. 1 is a flowchart of a fund recommendation process provided by an embodiment of the present invention. Figure 3 As shown in the figure, the workflow of fund recommendation is completed by the following components:

[0051] (1) Data acquisition device

[0052] Use a data collector to collect object data. You can obtain the relevant basic data of the object by establishing a user profile, and collect the marketing feedback information of the object to form the original data. Similar to predictive modeling, obtain the basic information of the object (such as age, gender, occupation, etc.), object behavior information (such as product preferences, channel preferences, transaction history, abnormal operations, risky transactions, etc.), object value information (such as whether it is a high-net-worth object, marketable potential, financial assets, etc.), and marketing feedback information (historical marketing records, push information click records, recommended page browsing time, outbound marketing call duration, etc.). After obtaining the above object data, integrate the data and enter the next stage of processing.

[0053] (2) Data processing devices

[0054] The data processor of the prediction unit also needs to process data in the same way as the optimal model generation unit. The data processor is used to fill in the missing data of the object. At the same time, feature extraction is performed on the object data. The object features are constructed and screened according to the feature columns selected in the optimal model generation unit. After being processed by the data processing device, the original object data can be processed into a feature data set that can be used for prediction.

[0055] (3) Amount prediction device

[0056] The amount prediction device is the optimal prediction model obtained through training above, which can predict the purchase amount range of the object for the fund through feature data. Different from the prediction of the object's purchase intention, the core of this application is to predict the purchase amount range of the object. The object feature data set processed by the data processing device is input into the prediction model, and the fund purchase amount range of the object is predicted by the prediction model. The obtained purchase amount range is divided according to the amount range in the interval marking unit, and the right endpoint of the interval is taken as the predicted amount. For example, if the predicted interval is 3,000 yuan to 4,000 yuan, then 4,000 yuan is taken as the predicted amount.

[0057] (4) Precision Marketing Devices

[0058] After obtaining the predicted amount of the fund purchased by the target, the precision marketing device can carry out targeted and differentiated precision marketing for the target based on the predicted purchase amount. The prediction results are handed over to the marketing personnel to formulate targeted marketing plans. According to the predicted amount, the corresponding fund message push, advertising marketing, etc. are sent to the target at regular intervals. SMS marketing or telephone marketing can also be carried out in traditional ways. Different from the existing solutions, the focus of this solution is to formulate marketing plans based on the amount range. Specifically, the marketing plan for the amount can adopt red envelope marketing (for example, buy 3,000 minus 20), cross-marketing of multiple categories of funds (for example, according to the predicted purchase amount of different funds by the target, carry out combined marketing of different categories), fee reduction marketing based on the amount (for example, reduction of agency fees), etc.

[0059] (5) Marketing effect feedback device

[0060] In addition, it is necessary to add an additional marketing effect feedback device. In the marketing result feedback device, information is collected for each marketing operation in the precision marketing device. For message push and advertising push, the user click push status and message browsing time can be collected. For telephone marketing, outbound call feedback information and call duration can be collected. The purchase behavior and amount of the object after each precision marketing are collected and compared with the predicted results. Targeted analysis is carried out based on the deviation of each purchase amount to discover new features. Ensure the data closed loop of the entire process of amount prediction-precision marketing-effect feedback, so as to carry out subsequent training and correction, and fully mine data information.

[0061] The technical solution provided by the embodiment of the present invention obtains the target fund recommendation model by obtaining a preliminary fund recommendation model and preset portrait sample information; training the preliminary fund recommendation model based on the preset portrait sample information; obtaining the portrait association information of the target object; inputting the portrait association information into the pre-trained target fund recommendation model to obtain the target recommendation result of the target object. The technical solution of the embodiment of the present invention solves the problem that the prior art only recommends fund products and the recommendation content is relatively single, and can predict the fund products and purchase amount requirements of the target object, thereby providing targeted services and promotions, improving user experience, and improving promotion effects.

[0062] Figure 4 It is a structural diagram of a fund recommendation device provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of recommending fund products to target objects. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.

[0063] like Figure 4As shown, the fund recommendation device includes: a road information acquisition module 310 , a control parameter determination module 320 and a control parameter determination module 330 .

[0064] Among them, the associated information acquisition module 310 is used to obtain the portrait associated information of the target object; the fund recommendation module 320 is used to input the portrait associated information into a pre-trained target fund recommendation model to obtain the target recommendation result of the target object; wherein the target fund recommendation model includes at least one target fund product and the target purchase amount range corresponding to the target fund product.

[0065] The technical solution provided by the embodiment of the present invention obtains the portrait association information of the target object; inputs the portrait association information into a pre-trained target fund recommendation model to obtain the target recommendation result of the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product. The technical solution of the embodiment of the present invention solves the problem that the prior art only recommends fund products and the recommendation content is relatively single, and can predict the fund product and purchase amount needs of the target object, thereby providing targeted services and promotions, improving user experience, and improving promotion effects.

[0066] In an optional embodiment, the target fund recommendation model includes a portrait feature extraction network and a portrait feature analysis network, and the fund recommendation module 320 is specifically used to: input the portrait association information into the portrait feature extraction network to obtain the target portrait feature of the target object; input the target portrait feature into the portrait feature analysis network to obtain the target recommendation result.

[0067] In an optional embodiment, the portrait feature analysis network includes: a fund recommendation sub-network and an amount recommendation sub-network, and the fund recommendation module 320 includes: a portrait feature analysis unit, which is used to: input the target portrait feature into the fund recommendation sub-network to obtain the at least one target fund product; input the target fund product and the target portrait feature into the amount recommendation sub-network to obtain the target purchase amount range corresponding to the target fund product.

[0068] In an optional implementation, the fund recommendation device further includes: a marketing recommendation module, which is used to: input the target recommendation result into the marketing recommendation network to obtain a marketing recommendation result corresponding to the target recommendation result.

[0069] In an optional embodiment, the fund recommendation device also includes: a model training module, which is used to: obtain a preliminary fund recommendation model and preset portrait sample information; train the preliminary fund recommendation model based on the preset portrait sample information to obtain the target fund recommendation model.

[0070] In an optional embodiment, the model training module includes: a model adjustment unit, used to: input the preset portrait sample information into the preliminary fund recommendation model to obtain a sample recommendation result; determine a recommendation loss function based on the sample recommendation result and a preset recommendation label of the preset portrait sample information; adjust the preliminary fund recommendation model according to the recommendation loss function to obtain the target fund recommendation model.

[0071] In an optional implementation, the target portrait features include: at least one of object basic features, object behavior features, object value features, and recommendation feedback features.

[0072] The fund recommendation device provided in the embodiment of the present invention can execute the fund recommendation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0073] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 5 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capability and can be configured in the fund recommendation device.

[0074] like Figure 5 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0075] The bus 18 may be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0076] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0077] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0078] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0079] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 20. Figure 5 As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 5Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0080] The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28, for example, implementing the fund recommendation method provided in the embodiment of the present invention, which includes:

[0081] Obtain portrait association information of the target object; input the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product.

[0082] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the fund recommendation method provided in any embodiment of the present invention is implemented, including:

[0083] Obtain portrait association information of the target object; input the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object; wherein the target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product.

[0084] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, 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), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0085] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0086] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0087] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or combinations thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0088] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0089] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A fund recommendation method, characterized in that: include: Obtain the image-related information of the target object; Inputting the portrait association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object; The target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product.

2. The method according to claim 1, characterized in that The target fund recommendation model includes a portrait feature extraction network and a portrait feature analysis network. The portrait association information is input into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object, including: Inputting the image association information into the image feature extraction network to obtain the target image feature of the target object; The target image features are input into the image feature analysis network to obtain the target recommendation result.

3. The method according to claim 2, characterized in that The image feature analysis network includes: a fund recommendation sub-network and an amount recommendation sub-network. The target image feature is input into the image feature analysis network to obtain the target recommendation result, including: Inputting the target profile feature into the fund recommendation sub-network to obtain the at least one target fund product; The target fund product and the target portrait features are input into the amount recommendation sub-network to obtain the target purchase amount range corresponding to the target fund product.

4. The method according to claim 1, characterized in that: The preliminary fund recommendation model also includes a promotion recommendation network, and the method further includes: The target recommendation result is input into the marketing recommendation network to obtain the marketing recommendation result corresponding to the target recommendation result.

5. The method according to claim 1, characterized in that The training process of the target fund recommendation model includes: Obtain preliminary fund recommendation model and preset profile sample information; The preliminary fund recommendation model is trained based on the preset portrait sample information to obtain the target fund recommendation model.

6. The method according to claim 5, characterized in that The training of the preliminary fund recommendation model based on the preset portrait sample information to obtain the target fund recommendation model includes: Inputting the preset portrait sample information into the preliminary fund recommendation model to obtain a sample recommendation result; Determine a recommendation loss function according to the sample recommendation result and the preset recommendation label of the preset portrait sample information; The preliminary fund recommendation model is adjusted according to the recommendation loss function to obtain the target fund recommendation model.

7. The method according to claim 2, characterized in that The target portrait features include: at least one of object basic features, object behavior features, object value features and recommendation feedback features.

8. A fund recommendation device, characterized in that: The device comprises: A related information acquisition module is used to obtain the image related information of the target object; A fund recommendation module, used for inputting the image association information into a pre-trained target fund recommendation model to obtain a target recommendation result for the target object; The target fund recommendation model includes at least one target fund product and a target purchase amount range corresponding to the target fund product.

9. A computer device, characterized in that: The computer device comprises: 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 fund recommendation method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the fund recommendation method as described in any one of claims 1 to 7 is implemented.

Images