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

Abstract

The invention discloses a financial product recommendation method and device, electronic equipment and a storage medium, and relates to the field of artificial intelligence, the field of financial science and technology or other related technical fields, and the recommendation method comprises the steps: obtaining a virtual scene set of a target user; each virtual scene in the virtual scene set is input to a feature recognition model one by one, a virtual scene feature set is output, and the virtual scene feature set at least comprises user image features and spatial features; inputting the virtual scene features in the virtual scene feature set into a classification model one by one, and outputting a financial preference tag set of the target user; and pushing a financial product to the target user based on the financial preference tags in the financial preference tag set. According to the method and the device, the technical problems of relatively single reference data type and relatively low recommendation accuracy of a financial product recommendation mode based on a user portrait in related technologies are solved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, financial technology or other related technical fields, and specifically, to a method and device for recommending financial products, an electronic device and a storage medium. Background Art

[0002] In recent years, with the continuous development of technologies such as artificial intelligence and cloud computing, the financial field has also kept pace with the times and adopted various new technologies to assist in the development of financial business. In order to enhance the stickiness between financial institutions and users, financial institutions will recommend suitable financial products to users based on user preferences, and plan personalized investment chains for users to improve user satisfaction.

[0003] In related technologies, with the rapid development of big data technology, it is proposed to obtain users' investment preferences by building user portraits and combining artificial intelligence technology, and recommend financial products to users based on their investment preferences. However, the financial product recommendation method based on user portraits has a relatively single reference data type, which is generally based on financial feature data extracted from the user's identity information and financial information, resulting in low accuracy of recommendations, which may be difficult to meet user needs. No effective solution has been proposed for the above problems. Summary of the invention

[0004] The embodiments of the present invention provide a method and device for recommending financial products, an electronic device and a storage medium, so as to at least solve the technical problem in the related art that the reference data type is relatively single and the recommendation accuracy is low in the method of recommending financial products based on user portraits.

[0005] According to one aspect of an embodiment of the present invention, there is provided a recommendation of a financial product, comprising: obtaining a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, where N is a positive integer; inputting each virtual scene in the virtual scene set into a feature recognition model one by one, and outputting a virtual scene feature set, wherein the virtual scene feature set includes at least: user image features and spatial features, and the feature recognition model is a pre-built model for extracting features; inputting virtual scene features in the virtual scene feature set into a classification model one by one, and outputting a financial preference tag set of the target user, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference tag set includes M financial preference tags, where M is a positive integer; and pushing financial products to the target user based on the financial preference tags in the financial preference tag set.

[0006] Optionally, after obtaining the virtual scene set of the target user, it also includes: mining the social relationship network of the target user based on the virtual scene corresponding to the target user; calculating the similarity between the target user and non-target users in the social relationship network; when the similarity between the target user and the non-target user is greater than a preset similarity threshold, obtaining the financial products purchased by the non-target user, and using the financial products purchased by the non-target user as alternative financial products to be pushed, and generating a list to be pushed.

[0007] Optionally, the step of pushing financial products to the target user based on the financial preference tags in the financial preference tag set includes: selecting a target financial product from the to-be-pushed list based on the financial preference tags; and pushing the target financial product to the target user.

[0008] Optionally, the virtual scene includes at least: a virtual image of the target user and a virtual space of the target user.

[0009] Optionally, the step of obtaining the virtual image of the target user includes: establishing a connection with a terminal device of the target user; obtaining user information of the target user based on the connection established with the target user; and constructing the virtual image based on the user information to obtain the virtual image of the target user.

[0010] Optionally, the step of obtaining the virtual space of the target user includes: obtaining the space where the target user is located and the spatial information corresponding to the space based on the connection established with the target user to obtain a spatial information set; counting the frequency of use of each space in the spatial information set by the target user; and obtaining the virtual space of the target user based on the use frequency.

[0011] Optionally, the step of obtaining the virtual space of the target user based on the usage frequency includes: comparing the usage frequency of each space by the target user with a preset usage frequency threshold, and when the usage frequency of the space is greater than or equal to the preset usage frequency threshold, treating the space as a high-frequency space of the target user; and constructing the virtual space based on the spatial information of the high-frequency space to obtain the virtual space of the target user.

[0012] Optionally, the step of constructing the classification model includes: obtaining a historical virtual scene set of the user and the historical financial preference label of the user within a historical time period; constructing a sample set based on the historical virtual scene set and the historical financial preference label; dividing the sample set based on a preset ratio to obtain a training set and a test set; iteratively training the initial model based on the training set to obtain an initial classification model; testing the initial classification model based on the test set to obtain a test result, and when the test result indicates that the initial classification model passes the test, the trained classification model is obtained.

[0013] According to another aspect of an embodiment of the present invention, a financial product recommendation device is also provided, comprising: an acquisition unit, used to acquire a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, and N is a positive integer; a first output unit, used to input each virtual scene in the virtual scene set into a feature recognition model one by one, and output a virtual scene feature set, wherein the virtual scene feature set includes at least: user image features and spatial features, and the feature recognition model is a pre-built model for extracting features; a second output unit, used to input the virtual scene features in the virtual scene feature set into a classification model one by one, and output a financial preference label set of the target user, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, and M is a positive integer; a push unit, used to push financial products to the target user based on the financial preference labels in the financial preference label set.

[0014] Optionally, the financial product recommendation device also includes: a first mining module, used to mine the social relationship network of the target user based on the virtual scene corresponding to the target user; a first calculation module, used to calculate the similarity between the target user and non-target users in the social relationship network; a first acquisition module, used to obtain the financial products purchased by the non-target user when the similarity between the target user and the non-target user is greater than a preset similarity threshold, and use the financial products purchased by the non-target user as alternative financial products to be pushed to generate a list to be pushed.

[0015] Optionally, the push unit includes: a first selection module, used to select a target financial product from the to-be-pushed list based on the financial preference tag; and a first push module, used to push the target financial product to the target user.

[0016] Optionally, the virtual scene includes at least: a virtual image of the target user and a virtual space of the target user.

[0017] Optionally, the acquisition unit includes: a first establishment module, used to establish a connection with the terminal device of the target user; a second acquisition module, used to acquire user information of the target user based on the connection established with the target user; and a first construction module, used to construct the virtual image based on the user information to obtain the virtual image of the target user.

[0018] Optionally, the acquisition unit also includes: a third acquisition module, used to establish a connection with the target user to acquire the space where the target user is located and the spatial information corresponding to the space, so as to obtain a spatial information set; a first statistical module, used to count the frequency of use of each space in the spatial information set by the target user; and a fourth acquisition module, used to acquire the virtual space of the target user based on the frequency of use.

[0019] Optionally, the fourth acquisition module includes: a first comparison submodule, used to compare the frequency of use of each space by the target user with a preset frequency of use threshold, and when the frequency of use of the space is greater than or equal to the preset frequency of use threshold, the space is regarded as a high-frequency space of the target user; a first construction submodule, used to construct the virtual space based on the spatial information of the high-frequency space to obtain the virtual space of the target user.

[0020] Optionally, the financial product recommendation device also includes: a sixth acquisition module, used to acquire a historical virtual scene set of the user and a historical financial preference label of the user within a historical time period; a second construction module, used to construct a sample set based on the historical virtual scene set and the historical financial preference label; a first division module, used to divide the sample set based on a preset ratio to obtain a training set and a test set; a first iteration module, used to iteratively train the initial model based on the training set to obtain an initial classification model; a first testing module, used to test the initial classification model based on the test set to obtain a test result, and when the test result indicates that the initial classification model passes the test, the trained classification model is obtained.

[0021] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for recommending financial products.

[0022] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising one or more processors and a memory, wherein the memory is used to store 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 any one of the above-mentioned methods for recommending financial products.

[0023] According to another aspect of an embodiment of the present invention, a computer program product is further provided, the computer program product comprising a computer program, wherein when the computer program is executed by a processor, any one of the above-mentioned methods for recommending financial products is implemented.

[0024] In the present application, the following steps are performed: first, a virtual scene set of a target user is obtained, wherein the virtual scene set includes N virtual scenes of the target user, where N is a positive integer; then, each virtual scene in the virtual scene set is input into a feature recognition model one by one, and a virtual scene feature set is output, wherein the virtual scene feature set includes at least: user image features and spatial features, and the feature recognition model is a pre-built model for extracting features; then, the virtual scene features in the virtual scene feature set are input into a classification model one by one, and a financial preference label set of the target user is output, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, where M is a positive integer; finally, financial products are pushed to the target user based on the financial preference labels in the financial preference label set.

[0025] In the present application, a virtual scene corresponding to a target user is obtained, and the virtual scene is input into a feature recognition model. The virtual scene features in the virtual scene are mined through the feature recognition model, and the virtual scene features output by the feature recognition model are input into a classification model. The financial preferences of the target user are identified based on the classification model to obtain multi-dimensional financial preference labels. Financial products are recommended to the target user through the financial preference labels. More potential feature information can be mined through the virtual scene of the target user, and the data sources obtained are more diversified, so the user's financial preferences can be analyzed more fully, thereby improving the accuracy of the recommendation results, making the financial products recommended to the user more in line with the user's preferences and needs, thereby solving the technical problem in the related technology that the method of recommending financial products based on user portraits has a relatively single reference data type and low recommendation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1is a flow chart of an optional method for recommending financial products according to an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram of an optional financial product recommendation process according to an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of an optional financial product recommendation device according to an embodiment of the present invention;

[0030] Figure 4 It is a hardware structure block diagram of an electronic device (or mobile device) for a method for recommending financial products according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only 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 should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] It should be noted that the financial product recommendation method and device in the present application can be used in the field of artificial intelligence to recommend financial products to target users based on the analysis of virtual scenes, and can also be used in any field other than the field of artificial intelligence to recommend financial products to target users based on the analysis of virtual scenes. The application field of the financial product recommendation method and device in the present application is not limited.

[0034] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0035] It should be noted that in this application, when collecting and analyzing customer information, corresponding operation entrances are provided for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered.

[0036] The following embodiments of the present invention can be applied to recommendation systems / applications / devices for various financial products. The present invention constructs a virtual person and a virtual space based on the target user, obtains a virtual scene corresponding to the target user, and mines virtual scene features from the virtual scene based on a feature recognition model, and inputs them into a classification model. The virtual scene features are analyzed by the classification model to determine the financial preference label of the target user, and then suitable financial products are recommended to the target user based on the financial preference of the target user, thereby improving user satisfaction and enhancing the stickiness between financial institutions and users.

[0037] The present invention is described in detail below in conjunction with various embodiments.

[0038] Embodiment 1

[0039] According to an embodiment of the present invention, an embodiment of a method for recommending financial products is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] Figure 1 is a flow chart of an optional method for recommending financial products according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0041] Step S101, obtaining a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, where N is a positive integer;

[0042] Step S102, input each virtual scene in the virtual scene set into a feature recognition model one by one, and output a virtual scene feature set, wherein the virtual scene feature set at least includes: user image features and space features, and the feature recognition model is a pre-built model for extracting features;

[0043] Step S103, inputting the virtual scene features in the virtual scene feature set into the classification model one by one, and outputting the target user's financial preference label set, wherein the classification model is a pre-trained model for classifying the target user's financial preferences, and the financial preference label set includes M financial preference labels, where M is a positive integer;

[0044] Step S104: Pushing financial products to target users based on the financial preference tags in the financial preference tag set.

[0045] Through the above steps, a virtual scene set of the target user is first obtained, wherein the virtual scene set includes N virtual scenes of the target user, N is a positive integer, and then each virtual scene in the virtual scene set is input into the feature recognition model one by one, and a virtual scene feature set is output, wherein the virtual scene feature set includes at least: user image features and space features, and the feature recognition model is a pre-built model for extracting features, and then the virtual scene features in the virtual scene feature set are input into the classification model one by one, and a financial preference label set of the target user is output, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, M is a positive integer, and finally, financial products are pushed to the target user based on the financial preference labels in the financial preference label set.

[0046] In this embodiment, a virtual scene corresponding to the target user is obtained, and the virtual scene is input into a feature recognition model. The virtual scene features in the virtual scene are mined through the feature recognition model, and the virtual scene features output by the feature recognition model are input into a classification model. The financial preferences of the target user are identified based on the classification model to obtain multi-dimensional financial preference labels. Financial products are recommended to the target user through the financial preference labels. More potential feature information can be mined through the virtual scene of the target user, and the data sources obtained are more diversified. The user's financial preferences can be analyzed more fully, thereby improving the accuracy of the recommendation results, making the financial products recommended to the user more in line with the user's preferences and needs, thereby solving the technical problem in the related art that the method of recommending financial products based on user portraits has a relatively single reference data type and low recommendation accuracy.

[0047] The embodiment of the present invention is described in detail below in combination with the above steps.

[0048] At present, with the continuous recovery of the economy, many users have the need to invest and manage their finances and purchase financial products. In order to recommend financial products to users more accurately, financial institutions use big data technology to mine users' preference information and recommend financial products that meet users' preferences based on the preference information.

[0049] It should be noted that the embodiment of the present invention is based on the metaverse technology to construct a virtual image for the target user, and combines the virtual image with the virtual space to construct a virtual scene corresponding to the target user, perform feature mining through the multi-dimensional information covered in the virtual scene, and analyze the mined features through a classification model to obtain the target user's financial preferences, thereby recommending financial products that meet the user's preferences and needs to the user.

[0050] Step S101, obtaining a virtual scene set of a target user.

[0051] Optionally, the virtual scene includes at least: a virtual image of the target user and a virtual space of the target user.

[0052] It should be noted that the above-mentioned virtual scene set includes N virtual scenes of the target user, where N is a positive integer. The virtual scene includes the virtual image and virtual space of the target user. The virtual image is a task model built based on metaverse technology, which can perform related actions and reflect the identity information of the target user. The virtual space refers to multiple occasions involving the target user, such as home space, work space, leisure space, entertainment space, etc.

[0053] Optionally, the step of obtaining the virtual image of the target user includes: establishing a connection with a terminal device of the target user; obtaining user information of the target user based on the connection established with the target user; and constructing a virtual image based on the user information to obtain the virtual image of the target user.

[0054] It should be noted that the virtual image of the target user is established based on the basic information and basic image of the target user. When constructing the virtual image of the target user, a connection is established with the terminal device of the target user (which may include mobile devices, PCs, tablets and other devices), and the user information of the target user is obtained through the established connection, such as name, age, occupation, clothing, appearance characteristics and other information, and then the virtual image of the target user is established based on the user information.

[0055] Optionally, the step of obtaining the virtual space of the target user includes: obtaining the space where the target user is located and the spatial information corresponding to the space based on the connection established with the target user to obtain a spatial information set; counting the frequency of use of each space in the spatial information set by the target user; and obtaining the virtual space of the target user based on the frequency of use.

[0056] It should be noted that the virtual space is constructed based on the occasions frequently visited by the target users. The geographical location, attributes and other information of the virtual space can reflect the characteristics of the users. When constructing the virtual space of the target users, the space and corresponding spatial information of the target users in the historical time period are first obtained based on the connection established with the terminal device to obtain a set of spatial information, and the frequency of use of each space by the target users is counted. The virtual scene to be constructed is determined based on the frequency of use.

[0057] Optionally, the step of obtaining the virtual space of the target user based on usage frequency includes: comparing the usage frequency of each space by the target user with a preset usage frequency threshold, and when the usage frequency of the space is greater than or equal to the preset usage frequency threshold, treating the space as a high-frequency space of the target user; and constructing a virtual space based on the spatial information of the high-frequency space to obtain the virtual space of the target user.

[0058] It should be noted that, according to the statistical results, if the target user's frequency of use of a certain space is greater than or equal to a preset frequency threshold, the space will be regarded as the target user's high-frequency space, and a virtual space will be established based on the high-frequency space. By combining the virtual image and the virtual space, the virtual scene of the target user can be obtained.

[0059] Optionally, after obtaining the virtual scene set of the target user, it also includes: mining the social relationship network of the target user based on the virtual scene corresponding to the target user; calculating the similarity between the target user and non-target users in the social relationship network; when the similarity between the target user and the non-target user is greater than a preset similarity threshold, obtaining the financial products purchased by the non-target user, and using the financial products purchased by the non-target user as alternative financial products to be pushed, and generating a list to be pushed.

[0060] It should be noted that after obtaining the virtual scene set of the target user, the social relationship network of the target user is mined from the virtual scene, and the similarity between the target user and the non-target users in the social relationship network is calculated. If the similarity is greater than the preset similarity threshold, the target user and the non-target user are determined to be similar users. The financial products that the user has purchased can be obtained from the similar users and used as alternative financial products to be recommended. A list to be pushed is generated to assist in the subsequent recommendation of financial products.

[0061] Step S102 , input each virtual scene in the virtual scene set into a feature recognition model one by one, and output a virtual scene feature set.

[0062] It should be noted that the virtual scenes are input into the feature recognition model one by one. The feature recognition model can extract relevant features from the virtual scene set to obtain a virtual scene feature set. The virtual scene feature set includes at least: user image features and spatial features. The feature recognition model is a pre-built model for extracting features.

[0063] Step S103, inputting the virtual scene features in the virtual scene feature set into the classification model one by one, and outputting the financial preference tag set of the target user.

[0064] It should be noted that after obtaining the virtual scene feature set, the virtual scene set is input into the classification model one by one, and the virtual scene features are analyzed and compared based on the classification model to determine the financial preferences of the target user, and a financial preference label set is output. The financial preference label set includes M financial preference labels, where M is a positive integer. The financial preference label set may include: investment intention label, risk preference label, financial product type preference label, investment duration preference label, etc. The investment intention label is used to indicate whether the target user has the demand and willingness for financial investment, the risk preference label is used to indicate whether the target user's risk preference when purchasing financial products is aggressive or conservative, the financial product type preference label is used to indicate the type of financial product preferred by the target user, such as stocks, funds, insurance, savings, etc., and the investment duration preference label is used to indicate whether the target user prefers long-term investment or short-term investment. The multi-dimensional financial preference labels can comprehensively reflect the investment preferences of the target user, and thus more accurately recommend financial products.

[0065] Optionally, the step of constructing a classification model includes: obtaining a historical virtual scene set of the user and the historical financial preference label of the user within a historical time period; constructing a sample set based on the historical virtual scene set and the historical financial preference label; dividing the sample set based on a preset ratio to obtain a training set and a test set; iteratively training the initial model based on the training set to obtain an initial classification model; testing the initial classification model based on the test set to obtain a test result, and when the test result indicates that the initial classification model passes the test, a trained classification model is obtained.

[0066] It should be noted that the classification model is a pre-trained model for classifying the financial preferences of target users. The steps of constructing the classification model include: obtaining a set of historical virtual scenes of users in a historical time period, and obtaining the historical financial preference labels of users, constructing a sample set based on the historical virtual scene set and the historical financial preference labels, and dividing the sample set into a training set and a test set. The training set is used to iteratively train the initial model to obtain an initial classification model, and the test set is used to test and optimize the initial classification model to obtain a trained classification model.

[0067] Step S104: Pushing financial products to target users based on the financial preference tags in the financial preference tag set.

[0068] It should be noted that the financial preferences of the target user can be determined based on the financial preference tags, and then suitable financial products can be pushed to the target user based on the user preferences. For example, suitable financial products can be screened from the financial product database based on the financial preference tags, and the product information of the financial product can be sent to the user terminal of the target user for the target user to select.

[0069] Optionally, the step of pushing financial products to target users based on the financial preference tags in the financial preference tag set includes: selecting a target financial product from a to-be-pushed list based on the financial preference tags; and pushing the target financial product to the target user.

[0070] It should be noted that when pushing financial products to target users, suitable financial products can also be screened from a list to be pushed that is pre-generated based on a social relationship network to obtain target financial products (the number of target financial products can be multiple), and the target financial products are pushed to the target user's device terminal for the user to select.

[0071] Another optional specific implementation is described in detail below.

[0072] Figure 2 is a schematic diagram of an optional financial product recommendation process according to an embodiment of the present invention, such as Figure 2 As shown in Figure 1, the process of recommending financial products to target users includes:

[0073] Step 1, start;

[0074] Step 2: Establish a connection with the user's device to obtain the user's virtual scene, wherein the virtual scene includes the user's virtual image and the virtual scene layout (ie, virtual space).

[0075] Step three, input the user's virtual scene into a feature recognition model, and output virtual scene features, which include virtual image features and virtual space features.

[0076] Step 4: Input the characteristics of the virtual scene into the classification model, and output the user's preference label (i.e., financial preference label) and the user's user characteristics, which can reflect the user's personalized characteristics, such as the user's interests and risk preference style. In particular, identify the user's characteristics related to financial services, such as whether he is willing to engage in investment-related financial activities, the liquidity of funds, whether he is willing to make long-term investments, etc., as well as whether his risk preference is aggressive or conservative, and whether he is more inclined to invest in stocks, funds, financial management, insurance, or savings.

[0077] Step 5: Recommend financial services (i.e., financial products) to users based on their preference tags. Users with the same preferences and characteristics are similar users, and the choices of other similar users can be used to recommend financial services to them.

[0078] Step six, end.

[0079] The embodiment of the present invention constructs a virtual person and a virtual space based on the target user, obtains a virtual scene corresponding to the target user, mines virtual scene features from the virtual scene based on a feature recognition model, and inputs the virtual scene features into a classification model. The virtual scene features are analyzed through the classification model to determine the financial preference label of the target user, and then suitable financial products are recommended to the target user based on the financial preference of the target user, thereby improving user satisfaction and enhancing the stickiness between financial institutions and users.

[0080] The following is a detailed description in conjunction with another embodiment.

[0081] Embodiment 2

[0082] A financial product recommendation device provided in this embodiment includes multiple implementation units, each of which corresponds to each implementation step in the above-mentioned embodiment 1. Its specific implementation methods and beneficial effects can refer to the above-mentioned method embodiments and will not be repeated here.

[0083] Figure 3 is a schematic diagram of an optional financial product recommendation device according to an embodiment of the present invention, such as Figure 2 As shown, the financial product recommendation device may include: an acquisition unit 31, a first output unit 32, a second output unit 33, and a push unit 34, wherein:

[0084] An acquisition unit 31 is used to acquire a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, where N is a positive integer;

[0085] The first output unit 32 is used to input each virtual scene in the virtual scene set into the feature recognition model one by one, and output the virtual scene feature set, wherein the virtual scene feature set at least includes: user image features and space features, and the feature recognition model is a pre-built model for extracting features;

[0086] A second output unit 33 is used to input the virtual scene features in the virtual scene feature set into the classification model one by one, and output the financial preference label set of the target user, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, where M is a positive integer;

[0087] The push unit 34 is used to push financial products to target users based on the financial preference tags in the financial preference tag set.

[0088] The above-mentioned financial product recommendation device can obtain a virtual scene set of the target user through an acquisition unit 31, wherein the virtual scene set includes N virtual scenes of the target user, and N is a positive integer; each virtual scene in the virtual scene set is input into a feature recognition model one by one through a first output unit 32, and a virtual scene feature set is output, wherein the virtual scene feature set includes at least: user image features and spatial features, and the feature recognition model is a pre-built model for extracting features; the virtual scene features in the virtual scene feature set are input into a classification model one by one through a second output unit 33, and a financial preference label set of the target user is output, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, and M is a positive integer; and the financial product is pushed to the target user based on the financial preference labels in the financial preference label set through a push unit 34.

[0089] In this embodiment, a virtual scene corresponding to the target user is obtained, and the virtual scene is input into a feature recognition model. The virtual scene features in the virtual scene are mined through the feature recognition model, and the virtual scene features output by the feature recognition model are input into a classification model. The financial preferences of the target user are identified based on the classification model to obtain multi-dimensional financial preference labels. Financial products are recommended to the target user through the financial preference labels. More potential feature information can be mined through the virtual scene of the target user, and the data sources obtained are more diversified. The user's financial preferences can be analyzed more fully, thereby improving the accuracy of the recommendation results, making the financial products recommended to the user more in line with the user's preferences and needs, thereby solving the technical problem in the related art that the method of recommending financial products based on user portraits has a relatively single reference data type and low recommendation accuracy.

[0090] Optionally, the financial product recommendation device also includes: a first mining module, used to mine the social relationship network of the target user based on the virtual scene corresponding to the target user; a first calculation module, used to calculate the similarity between the target user and the non-target users in the social relationship network; a first acquisition module, used to obtain the financial products purchased by the non-target user when the similarity between the target user and the non-target user is greater than a preset similarity threshold, and use the financial products purchased by the non-target user as alternative financial products to be pushed to generate a list to be pushed.

[0091] Optionally, the push unit includes: a first selection module, used to select a target financial product from a to-be-pushed list based on the financial preference tag; and a first push module, used to push the target financial product to a target user.

[0092] Optionally, the virtual scene includes at least: a virtual image of the target user and a virtual space of the target user.

[0093] Optionally, the acquisition unit includes: a first establishment module, used to establish a connection with a terminal device of a target user; a second acquisition module, used to acquire user information of the target user based on the connection established with the target user; and a first construction module, used to construct a virtual image based on the user information to obtain the virtual image of the target user.

[0094] Optionally, the acquisition unit also includes: a third acquisition module, which is used to establish a connection with the target user to acquire the space where the target user is located and the spatial information corresponding to the space, and obtain a spatial information set; a first statistical module, which is used to count the frequency of use of each space in the spatial information set by the target user; and a fourth acquisition module, which is used to acquire the virtual space of the target user based on the frequency of use.

[0095] Optionally, the fourth acquisition module includes: a first comparison submodule, used to compare the frequency of use of each space by the target user with a preset frequency of use threshold, and when the frequency of use of the space is greater than or equal to the preset frequency of use threshold, the space is regarded as a high-frequency space of the target user; a first construction submodule, used to construct a virtual space based on the spatial information of the high-frequency space to obtain the virtual space of the target user.

[0096] Optionally, the financial product recommendation device also includes: a sixth acquisition module, used to obtain the user's historical virtual scene set and the user's historical financial preference label within a historical time period; a second construction module, used to construct a sample set based on the historical virtual scene set and the historical financial preference label; a first division module, used to divide the sample set based on a preset ratio to obtain a training set and a test set; a first iteration module, used to iteratively train the initial model based on the training set to obtain an initial classification model; a first testing module, used to test the initial classification model based on the test set to obtain a test result, and when the test result indicates that the initial classification model passes the test, a trained classification model is obtained.

[0097] The financial product recommendation device may further include a processor and a memory. The acquisition unit 31, the first output unit 32, the second output unit 33, the push unit 34, etc. are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions.

[0098] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and financial products can be recommended to users by adjusting kernel parameters.

[0099] The above-mentioned memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (fl ash RAM), and the memory includes at least one storage chip.

[0100] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for recommending financial products.

[0101] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors implement any one of the above-mentioned methods for recommending financial products.

[0102] According to another aspect of an embodiment of the present invention, a computer program product is provided. The computer program product includes a computer program, wherein when the computer program is executed by a processor, any one of the above-mentioned methods for recommending financial products is implemented.

[0103] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: obtaining a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, and N is a positive integer; inputting each virtual scene in the virtual scene set into a feature recognition model one by one, and outputting a virtual scene feature set, wherein the virtual scene feature set includes at least: user image features and spatial features, and the feature recognition model is a pre-built model for extracting features; inputting the virtual scene features in the virtual scene feature set into a classification model one by one, and outputting a financial preference label set of the target user, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, and M is a positive integer; and pushing financial products to the target user based on the financial preference labels in the financial preference label set.

[0104] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that initializes the following method steps: after obtaining a set of virtual scenes for a target user, it also includes: mining a social relationship network of the target user based on the virtual scenes corresponding to the target user; calculating the similarity between the target user and non-target users in the social relationship network; when the similarity between the target user and the non-target user is greater than a preset similarity threshold, obtaining a financial product purchased by the non-target user, and using the financial product purchased by the non-target user as an alternative financial product to be pushed, and generating a list to be pushed.

[0105] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that initializes the following method steps: the step of pushing financial products to target users based on financial preference tags in a financial preference tag set includes: selecting a target financial product from a list to be pushed based on the financial preference tag; and pushing the target financial product to the target user.

[0106] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that initializes the following method steps: the virtual scene at least includes: a virtual image of a target user and a virtual space of the target user.

[0107] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that initializes the following method steps: the steps of obtaining a virtual image of a target user include: establishing a connection with a terminal device of the target user; obtaining user information of the target user based on the connection established with the target user; constructing a virtual image based on the user information to obtain a virtual image of the target user.

[0108] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that initializes the following method steps: the step of obtaining the virtual space of the target user includes: obtaining the space where the target user is located and the spatial information corresponding to the space based on the connection established with the target user, and obtaining a spatial information set; counting the frequency of use of each space in the spatial information set by the target user; and obtaining the virtual space of the target user based on the frequency of use.

[0109] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that initializes the following method steps: the step of obtaining the virtual space of the target user based on the frequency of use includes: comparing the frequency of use of each space by the target user with a preset frequency of use threshold, and when the frequency of use of the space is greater than or equal to the preset frequency of use threshold, treating the space as a high-frequency space of the target user; constructing a virtual space based on the spatial information of the high-frequency space to obtain the virtual space of the target user.

[0110] The present application also provides a computer program product, which, when executed on a data processing device, is also suitable for executing a program that is initialized with the following method steps: the step of constructing a classification model includes: obtaining a historical virtual scene set of users and the financial preference labels of users within a historical time period; constructing a sample set based on the historical virtual scene set and the financial preference labels; dividing the sample set based on a preset ratio to obtain a training set and a test set; iteratively training the initial model based on the training set to obtain an initial classification model; testing the initial classification model based on the test set to obtain a test result, and when the test result indicates that the initial classification model passes the test, a trained classification model is obtained.

[0111] Figure 4 1 is a hardware structure block diagram of an electronic device (or mobile device) for a method for recommending financial products according to an embodiment of the present invention. Figure 4 As shown, the electronic device may include one or more ( Figure 4 402a, 402b, ..., 402n are used to illustrate) processor 402 (processor 402 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), memory 404 for storing data. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a keyboard, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 4 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 4 More or fewer components as shown, or with Figure 4 Different configurations shown.

[0112] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0113] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0114] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0115] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0116] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0117] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0118] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for recommending financial products, characterized in that: include: Acquire a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, where N is a positive integer; Input each virtual scene in the virtual scene set into a feature recognition model one by one, and output a virtual scene feature set, wherein the virtual scene feature set at least includes: user image features and space features, and the feature recognition model is a pre-built model for extracting features; Inputting the virtual scene features in the virtual scene feature set into the classification model one by one, and outputting the financial preference label set of the target user, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, where M is a positive integer; Financial products are pushed to the target user based on the financial preference tags in the financial preference tag set.

2. The method according to claim 1, characterized in that After obtaining the target user's virtual scene set, it also includes: Mining the social relationship network of the target user based on the virtual scene corresponding to the target user; Calculating the similarity between the target user and non-target users in the social relationship network; When the similarity between the target user and the non-target user is greater than a preset similarity threshold, the financial products purchased by the non-target user are obtained, and the financial products purchased by the non-target user are used as candidate financial products to be pushed to generate a list to be pushed.

3. The method according to claim 2, characterized in that The step of pushing financial products to the target user based on the financial preference tags in the financial preference tag set includes: Selecting a target financial product from the list to be pushed based on the financial preference tag; Pushing the target financial product to the target user.

4. The method according to claim 1, characterized in that: The virtual scene includes at least: a virtual image of the target user and a virtual space of the target user.

5. The method according to claim 4, characterized in that The step of obtaining the virtual image of the target user includes: Establishing a connection with a terminal device of the target user; Acquire user information of the target user based on the connection established with the target user; The virtual image is constructed based on the user information to obtain the virtual image of the target user.

6. The method according to claim 4, characterized in that The step of acquiring the virtual space of the target user includes: Acquire the space where the target user is located and the spatial information corresponding to the space based on the connection established with the target user, and obtain a spatial information set; Counting the usage frequency of each space in the space information set by the target user; The virtual space of the target user is acquired based on the usage frequency.

7. The method according to claim 6, characterized in that The step of acquiring the virtual space of the target user based on the usage frequency includes: Comparing the usage frequency of each space by the target user with a preset usage frequency threshold, and if the usage frequency of a space is greater than or equal to the preset usage frequency threshold, treating the space as a high-frequency space of the target user; The virtual space is constructed based on the spatial information of the high-frequency space to obtain the virtual space of the target user.

8. The method according to claim 1, characterized in that The steps of constructing the classification model include: Obtain the user's historical virtual scene set and the user's historical financial preference label within the historical time period; Constructing a sample set based on the historical virtual scene set and the historical financial preference label; Dividing the sample set based on a preset ratio to obtain a training set and a test set; Iteratively training the initial model based on the training set to obtain an initial classification model; The initial classification model is tested based on the test set to obtain a test result. When the test result indicates that the initial classification model passes the test, the trained classification model is obtained.

9. A financial product recommendation device, characterized in that: include: An acquisition unit, configured to acquire a virtual scene set of a target user, wherein the virtual scene set includes N virtual scenes of the target user, where N is a positive integer; A first output unit is used to input each virtual scene in the virtual scene set into a feature recognition model one by one, and output a virtual scene feature set, wherein the virtual scene feature set at least includes: user image features and space features, and the feature recognition model is a pre-built model for extracting features; A second output unit is used to input the virtual scene features in the virtual scene feature set into a classification model one by one, and output a financial preference label set of the target user, wherein the classification model is a pre-trained model for classifying the financial preferences of the target user, and the financial preference label set includes M financial preference labels, where M is a positive integer; A pushing unit is used to push financial products to the target user based on the financial preference tags in the financial preference tag set.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the financial product recommendation method according to any one of claims 1 to 8.

11. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store 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 financial product recommendation method described in any one of claims 1 to 8.

12. A computer program product, characterized in that The computer program product comprises a computer program, wherein the computer program, when executed by a processor, implements the method for recommending financial products according to any one of claims 1 to 8.

Images