Training of product recommendation model, product recommendation method, device and electronic equipment

Abstract

The present disclosure provides a product recommendation model training method, a product recommendation method, a device and an electronic device, which can be applied to the technical field of big data, artificial intelligence and the like, and the field of financial technology. The method comprises: obtaining a product relationship graph model constructed according to product labels, wherein the product relationship graph model comprises first node information, second node information and edge information, the first node information comprises a first product label and a first user label, the second node information only comprises a second product label, and the edge information represents that two nodes corresponding to the edge information have an association relationship; determining a second user label corresponding to the second product label according to the first product label, the first user label, the second product label and the edge information; and obtaining a trained product recommendation model according to the first node information, the edge information and the second node information having the second product label and the second user label.

Description

Technical Field

[0001] This disclosure relates to the fields of big data, artificial intelligence and other technologies, as well as the field of fintech, and in particular to the training of a product recommendation model, a product recommendation method, an apparatus and an electronic device. Background Technology

[0002] Currently, banks offer a wide variety of investment products, including self-operated loan products and financial market products, as well as fund products, wealth management products, and insurance products sold on behalf of other institutions. Each type of investment product has a large number of options, and each type of product has its own recommendation model to help ordinary users find products that meet their needs.

[0003] In the process of realizing the present invention, the inventors discovered that this recommendation model based on product type has a drawback: customers need to enter the actual product page to see recommendations for a single product category, resulting in a poor user experience.

[0004] Therefore, how to provide customers with accurate and comprehensive product profiles and recommendations based on the numerous types and vast amount of investment products is a problem that the industry urgently needs to solve. Summary of the Invention

[0005] In view of the above problems, this disclosure provides training of product recommendation models, product recommendation methods, apparatus and electronic devices.

[0006] According to one aspect of this disclosure, a method for training a product recommendation model is provided, comprising: obtaining a product relationship graph model constructed based on product tags, wherein the product relationship graph model includes first node information, second node information, and edge information, the first node information including a first product tag and a first user tag, the second node information including only a second product tag, and the edge information representing an association relationship between two nodes corresponding to the edge information; determining a second user tag corresponding to the second product tag based on the first product tag, the first user tag, the second product tag, and the edge information; and obtaining a trained product recommendation model based on the first node information, the edge information, and the second node information having the second product tag and the second user tag.

[0007] According to another aspect of this disclosure, a product recommendation method is provided, comprising: determining target user tags based on target user attribute information of a target user; and inputting the target user tags into a product recommendation model to obtain target products recommended for the target user, wherein the product recommendation model is a trained product recommendation model obtained according to the training method of the product recommendation model described in this disclosure.

[0008] According to another aspect of this disclosure, a training apparatus for a product recommendation model is provided, comprising: a first acquisition module, configured to acquire a product relationship graph model constructed based on product tags, wherein the product relationship graph model includes first node information, second node information, and edge information, the first node information including a first product tag and a first user tag, the second node information including only a second product tag, and the edge information representing an association relationship between two nodes corresponding to the edge information; a first determination module, configured to determine a second user tag corresponding to the second product tag based on the first product tag, the first user tag, the second product tag, and the edge information; and a first acquisition module, configured to obtain a trained product recommendation model based on the first node information, the edge information, and the second node information having the second product tag and the second user tag.

[0009] According to another aspect of this disclosure, a product recommendation device is provided, comprising: a fifth determining module, configured to determine target user tags based on target user attribute information of a target user; and a second obtaining module, configured to input the target user tags into a product recommendation model to obtain target products recommended for the target user, wherein the product recommendation model is a trained product recommendation model obtained according to the product recommendation model training device of this disclosure.

[0010] According to another aspect of this disclosure, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform at least one of the training method for a product recommendation model and the product recommendation method described in this disclosure.

[0011] According to another aspect of this disclosure, a computer-readable storage medium is also provided, having stored thereon executable instructions that, when executed by a processor, cause the processor to perform at least one of the training method for the product recommendation model and the product recommendation method described in this disclosure.

[0012] According to another aspect of this disclosure, a computer program product is also provided, including a computer program that, when executed by a processor, implements at least one of the training method and product recommendation method of the product recommendation model described in this disclosure.

[0013] According to the product recommendation model training, product recommendation method, apparatus, and electronic device provided in this disclosure, a product relationship graph model constructed based on product tags is obtained. This model includes first node information, second node information, and edge information. The first node information includes a first product tag and a first user tag, the second node information only includes a second product tag, and the edge information represents the association between two nodes corresponding to the edge information. Based on the first product tag, the first user tag, the second product tag, and the edge information, a second user tag corresponding to the second product tag is determined. Finally, based on the first node information, the edge information, and the second node information containing both the second product tag and the second user tag, a trained product recommendation model is obtained. Since product nodes can be labeled based on user tags, and a label propagation algorithm can be used to update the label information of unlabeled product nodes using labeled product nodes, propagating throughout the network until convergence, a community label is formed based on the basic information of product tags and user tags. For users matching the tags, various types of products within the corresponding community labels are recommended, at least partially alleviating the technical problem of needing to enter the actual product page to see recommendations for a single product category. This technology enables the creation of comprehensive product profiles and recommendations based on customers' user characteristics, providing users with all-round investment guidance. Attached Figure Description

[0014] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0015] Figure 1 The illustration schematically shows an application scenario of at least one of the training methods and product recommendation methods of the product recommendation model according to embodiments of the present disclosure;

[0016] Figure 2 A flowchart illustrating a method for training a product recommendation model according to an embodiment of the present disclosure is shown schematically.

[0017] Figure 3 A schematic diagram of a product relationship similarity matrix model according to an embodiment of the present disclosure is shown.

[0018] Figure 4 A schematic diagram of an initial product relationship diagram model according to an embodiment of the present disclosure is shown.

[0019] Figure 5 A flowchart illustrating the training of a product relationship graph model according to an embodiment of the present disclosure is shown schematically;

[0020] Figure 6 A schematic diagram of a trained product relationship graph model according to an embodiment of the present disclosure is shown.

[0021] Figure 7 A flowchart illustrating a product recommendation method based on the trained product recommendation model described above, according to an embodiment of this disclosure, is shown in the schematic diagram.

[0022] Figure 8 This schematic diagram illustrates a structural block diagram of a training apparatus for a product recommendation model according to an embodiment of the present disclosure;

[0023] Figure 9 A schematic block diagram of a product recommendation device according to an embodiment of the present disclosure is shown; and

[0024] Figure 10 A block diagram of an electronic device suitable for implementing at least one of the training methods and product recommendation methods of a product recommendation model, according to embodiments of the present disclosure, is illustrated. Detailed Implementation

[0025] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0026] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

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

[0028] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0029] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision, disclosure, and application of data (including but not limited to user personal information) comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.

[0030] Embodiments of this disclosure provide a product recommendation model training method, apparatus, and electronic device. The product recommendation model training method includes: obtaining a product relationship graph model constructed based on product tags, wherein the product relationship graph model includes first node information, second node information, and edge information; the first node information includes a first product tag and a first user tag; the second node information includes only a second product tag; and the edge information represents the association relationship between two nodes corresponding to the edge information; determining a second user tag corresponding to the second product tag based on the first product tag, the first user tag, the second product tag, and the edge information; and obtaining a trained product recommendation model based on the first node information, the edge information, and the second node information having the second product tag and the second user tag.

[0031] Figure 1 The illustration schematically depicts an application scenario of at least one of the training methods and product recommendation methods for a product recommendation model according to embodiments of the present disclosure.

[0032] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0033] Users can interact with server 105 via network 104 using at least one of the first terminal device 101, second terminal device 102, and third terminal device 103 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, second terminal device 102, and third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0034] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0035] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0036] It should be noted that at least one of the training methods and product recommendation methods of the product recommendation model provided in this disclosure embodiment can generally be executed by server 105. Correspondingly, at least one of the training devices and product recommendation devices of the product recommendation model provided in this disclosure embodiment can generally be located in server 105. At least one of the training methods and product recommendation methods of the product recommendation model provided in this disclosure embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, at least one of the training devices and product recommendation devices of the product recommendation model provided in this disclosure embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0037] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0038] It should be noted that the training, product recommendation method, device, and electronic equipment of the disclosed product recommendation model can be used in the fields of big data, artificial intelligence, and fintech, and can also be used in any field other than big data, artificial intelligence, and fintech. The application fields of the training, product recommendation method, device, and electronic equipment of the disclosed product recommendation model are not limited.

[0039] The following will be based on Figure 1 The described scene, through Figures 2-6 The training method of the product recommendation model of the disclosed embodiments is described in detail.

[0040] Figure 2 A flowchart illustrating a method for training a product recommendation model according to an embodiment of the present disclosure is shown.

[0041] like Figure 2 As shown, the training method of the product recommendation model in this embodiment includes operations S210 to S230.

[0042] In operation S210, a product relationship graph model constructed based on product tags is obtained. The product relationship graph model includes first node information, second node information, and edge information. The first node information includes the first product tag and the first user tag. The second node information only includes the second product tag. The edge information represents the association relationship between the two nodes corresponding to the edge information.

[0043] According to embodiments of this disclosure, products may include various multimedia information, such as news, novels, and videos, as well as various financial products, such as funds, deposits, and investment products, and are not limited to these. For example, a product may be a financial product, and product tags (including a first product tag and a second product tag) may include at least one of the following: product type, number of purchases, number of redemptions, yield, total yield, net asset value, risk index, etc., and are not limited to these. Data acquisition tools can be used to obtain these feature values ​​for each financial product and construct a data entry to obtain product tags. Where a certain type of product does not involve certain features, it can be marked as not involving them.

[0044] For example, based on attributes such as product type, number of purchases, number of redemptions, yield, total yield, net asset value, and risk index, the following data can be obtained for investment product A: Fund, 1,000,000, 500,000, 3.00%, 10.55%, 1.5502, medium risk; and the following data can be obtained for investment product B: Deposit, 1,000,000, not involved, 2.45%, 2.45%, 1.00, low risk.

[0045] According to embodiments of this disclosure, the product relationship graph model can be a model with a graph structure. The graph structure can include nodes and edges. Each node in the product relationship graph model can correspond to a product, including first node information or second node information corresponding to that product. Each edge in the product relationship graph model can represent edge information.

[0046] According to embodiments of this disclosure, a product relationship diagram can be determined based on a first product tag and a second product tag. Then, a first user tag can be configured for one or more nodes in the product relationship diagram to obtain an initial product relationship diagram model.

[0047] In operation S220, based on the first product label, the first user label, the second product label, and the edge information, the second user label corresponding to the second product label is determined.

[0048] According to embodiments of this disclosure, both the first user tag and the second user tag may include at least one of the following: age, gender, income, education level, occupation, risk tolerance, customer level, etc., and are not limited to these. Data acquisition tools can be used to obtain these characteristic values ​​for each individual user or each product's existing historical users, constructing a data entry to obtain the user tag. Risk tolerance can be obtained through surveys and analysis of users during their initial product purchases.

[0049] It should be noted that the data acquisition tools mentioned above may include, but are not limited to, web crawlers.

[0050] According to embodiments of this disclosure, by using a first user tag as input to a product relationship graph model and combining it with LPA (Label Propagation Algorithm), the initial product relationship graph model is trained, for example, to obtain a second user tag that matches each second product tag.

[0051] In operation S230, a trained product recommendation model is obtained based on the first node information, edge information, and second node information with second product label and second user label.

[0052] According to embodiments of this disclosure, each node in the trained product recommendation model may include a product tag (including a first product tag or a second product tag) and a user tag (including a first user tag or a second user tag) that is adapted to the product tag.

[0053] Through the embodiments of this disclosure, product nodes can be labeled based on user tags, and the tag propagation algorithm can be used to update the tag information of unlabeled product nodes using product nodes already labeled with user tags, propagating throughout the network until convergence, forming community tags determined by the basic information of product tags and user tags. For users matching the tags, various types of products within the corresponding community tags are recommended, at least partially alleviating the technical problem of needing to enter the actual product page to see recommendations for a single product category. This achieves the technical effect of obtaining a three-dimensional product profile and product recommendations based on the customer's own user characteristic tags, providing users with comprehensive investment guidance.

[0054] The following describes specific embodiments. Figure 2 The method shown will be explained in further detail.

[0055] According to embodiments of this disclosure, before performing the above-described operation S210, an initial product relationship graph model can be constructed first. This construction method may include: obtaining first product attribute information of a first product and second product attribute information of a second product; performing a structured transformation on the first product attribute information according to a first predefined mapping relationship to obtain a first product tag; wherein the first predefined mapping relationship represents the mapping relationship between various types of product attribute information and numerical information; performing a structured transformation on the second product attribute information according to the first predefined mapping relationship to obtain a second product tag; obtaining user attribute information of users related to the first product; performing a structured transformation on the user attribute information according to a second predefined mapping relationship to obtain a first user tag; wherein the second predefined mapping relationship represents the mapping relationship between various types of user attribute information and numerical information; determining first node information based on the first product tag and the first user tag corresponding to the same first product; determining second node information based on the second product tag; and determining edge information based on the first similarity between every two product tags in the first and second product tags.

[0056] According to embodiments of this disclosure, the first product attribute information and the second product attribute information may include at least one of the following: product type, number of purchases, number of redemptions, yield, total yield, net asset value, risk index, etc., and are not limited to these. The first product tag and the second product tag may also represent feature values ​​of the aforementioned product attribute information. The first predefined mapping relationship may include the correspondence between various product attribute information and their feature values, and is not limited to these. For example, the first predefined mapping relationship may include the content shown in Tables 1 and 2. Table 1 may define the correspondence between a product involving / not involving this tag and dictionary values. Table 2 may define the correspondence between product type and dictionary values.

[0057] Table 1:

[0058] Does this label apply? dictionary value Involving 1 Not involved 0

[0059] Table 2:

[0060] Investment product types dictionary value fund 1 Financial Management 2 Insurance 3 deposit 4 ... ...

[0061] According to embodiments of this disclosure, structured transformation can, for example, represent the process of converting textual information into dictionary values. For instance, by performing structured transformation on the obtained first product attribute information and second product attribute information in conjunction with the aforementioned Tables 1 and 2, a product label table as shown in Table 3 can be obtained.

[0062] Table 3:

[0063]

[0064] According to embodiments of this disclosure, after obtaining a digital first product label and a second product, a process can be performed to determine edge information based on a first similarity between every two product labels in the first and second product labels. After determining the edge information, a product relationship similarity matrix model can be constructed, for example.

[0065] Figure 3 A schematic diagram of a product relationship similarity matrix model according to an embodiment of the present disclosure is shown.

[0066] like Figure 3 As shown, the product relationship similarity matrix model includes nodes a, b, c, d, e, f, g, h, i, and edges ab, ac, bc, ad, ce, ef, ei, eg, ef, fh, fi, gh, hi. Each node includes a product label corresponding to the product (including the first product label and the second product label mentioned above).

[0067] According to embodiments of this disclosure, user attribute information may include at least one of the following: age, gender, income, education level, occupation, risk tolerance, customer level, etc., and is not limited to these. The first user tag may also be represented as the feature value of the aforementioned user attribute information. The second predefined mapping relationship may include the correspondence between various user attribute information and their feature values, and is not limited to these. For example, the second predefined mapping relationship may include the content shown in Table 4. Table 4 can define a gender data dictionary.

[0068] Table 4:

[0069]

[0070] According to embodiments of this disclosure, gender, risk tolerance, occupation, and education level in user attribute information are unstructured data. Before modeling, a data dictionary can be established based on a second predefined mapping relationship to convert them into digital structured data. For example, after structural conversion of the obtained user attribute information, an existing investment user information table as shown in Table 5 can be obtained.

[0071] Table 5:

[0072]

[0073] According to embodiments of this disclosure, the first user tag can also be determined by combining one or more user characteristics with an existing user tagging system. For example, an existing user tagging system may include customer group tags (18-30 years old: young customer group; 31-40 years old: middle-aged customer group; 45-60 years old: middle-aged customer group), risk tolerance tags (1-2: low risk, 3-4: medium risk, 5: high risk), customer quality tags (income > 50 and education level 5: excellent, 50 > income > 35 and education level 4: average), and is not limited to these. Based on this existing user tagging system, tagging the users in Table 5 can, for example, yield the results shown in Table 6.

[0074] Table 6:

[0075] user Customer tags Risk tolerance label Customer quality label User A Adult customers Medium risk medium User B young customers Medium risk high quality ... ... ... ...

[0076] According to embodiments of this disclosure, after obtaining the user attribute information as shown in Table 6, the unstructured data in Table 6 can be structured by combining a second predefined mapping relationship to obtain a numerical first user tag. Then, the first user tag can be used as input parameters into an LPA-based product relationship graph model to label some nodes in the product relationship graph model, for example, to obtain an initial product relationship graph model.

[0077] Figure 4 A schematic diagram of an initial product relationship diagram model according to an embodiment of the present disclosure is shown.

[0078] like Figure 4 As shown, the initial product relationship graph model includes nodes a, b, c, d, e, f, g, h, and i, and edges ab, ac, bc, ad, ce, ef, ei, eg, ef, fh, fi, gh, and hi. Node a is configured with a first user label, such as "young customer group," but this is not limited to these. Node h is configured with a first user label, such as "middle-aged customer group," but this is not limited to these.

[0079] Through the embodiments described above in this disclosure, combined with structured processing, data in model calculations can be converted into a unified data format, facilitating model processing and improving model accuracy.

[0080] According to embodiments of this disclosure, the above operation S220 may include: determining the information propagation probability between the first node information and the second node information based on a second similarity between the first product tag and the second product tag; and determining the second user tag based on the first user tag and the information propagation probability.

[0081] According to embodiments of this disclosure, the second similarity can be determined by calculating the Euclidean distance between the first product label and the second product label. The second similarity can be determined as the information propagation probability, or it can be further processed using a predefined formula to obtain the information propagation probability.

[0082] According to embodiments of this disclosure, determining the information propagation probability between the first node information and the second node information based on the second similarity between the first product label and the second product label may include: determining first edge weight information between the first node information and the second node information based on preset constraint parameters and the Euclidean distance between the first product label and the second product label; determining second edge weight information between every two node information in the first node information and the second node information; and determining the information propagation probability based on the first edge weight information and the second edge weight information.

[0083] According to embodiments of this disclosure, edge weight information can characterize the weight of edges in a product relationship graph model. The larger the edge weight, the more similar the two nodes are, and the easier it is for user tags on one node to propagate to another node.

[0084] For example, the edge weight information (including the first edge weight information and the second edge weight information mentioned above) can be determined by combining formula (1).

[0085]

[0086] In formula (1), w ij It can represent the weight information of the first edge between node i and node j, x i x j It can represent the first product label and the second product label respectively, or it can represent the product label corresponding to every two nodes in the first node information and the second node information respectively. α can represent the preset constraint parameter.

[0087] For example, the probability of information propagation can be determined by combining formula (2).

[0088]

[0089] In formula (2), P ij w can represent the probability of information propagation from node i to node j. ik It can represent the weight information of the second edge between every two nodes in the first node information and the second node information. n can represent the number of nodes, and i, j, k can represent the parameters representing the nodes.

[0090] It should be noted that the above method of determining the second similarity based on Euclidean distance, determining the edge weight information based on formula (1), and determining the information propagation probability based on formula (2) is only an exemplary embodiment. In actual implementation, it is not limited to this method. Any one of these methods can determine the second similarity, edge weight information, and information propagation probability.

[0091] According to embodiments of this disclosure, a second user tag can be determined based on a first user tag when the probability of information propagation is determined to be greater than a preset probability value. In this process, for example, the first user tag can be used as the second user tag. Alternatively, the first user tag and the probability value represented by the probability of information propagation can be used as the second user tag.

[0092] Through the above embodiments of this disclosure, user tags corresponding to each product tag can be calculated relatively accurately based on the tag propagation algorithm, and the trained model can have high output accuracy.

[0093] According to embodiments of this disclosure, operation S220 may further include: determining a probability transition matrix based on the information propagation probability between every two nodes in the first node information and the second node information; determining an initial user soft tag matrix based on the first user tag and initial parameter information representing the second user tag; determining a target user soft tag matrix based on the probability transition matrix and the initial user soft tag matrix, wherein the elements in the target user soft tag matrix include user tag information and probability values ​​corresponding to the user tag information; and determining the second user tag based on the target user soft tag matrix.

[0094] For example, let (x1, y1)…(x1, y1) define the first node information. Here, XL = {x1…x1} is the first product label, YL = {y1…y1}∈{1…C} is the first user label, the number of labels C is known, and all exist in the label data. Let (x1+1, y1+1)…(x1+u, y1+u) define the second node information. Here, XU = {x1+1…x1+u} is the second product label, YU = {y1+1…y1+u} is the second user label to be predicted, 1 << u, N = 1+U. Let the dataset X = {x1…x1+u}∈R. The training process of the product relationship graph model can be transformed into: from the dataset X, using the learning of YL, predicting the corresponding second user label for each second product label in the unlabeled dataset YU.

[0095] According to embodiments of this disclosure, for example, the aforementioned formulas (1) to (2) can be combined to calculate the information propagation probability for every two nodes in the first node information and the second node information. For example, corresponding to the above embodiments, by calculating the information propagation probability between every two nodes, an N×N probability transition matrix P can be obtained. Combining YL and YU, an N×C initial user soft label matrix F = [YL; YU] can be obtained. By performing a training process on F for multiple rounds according to P, the target user soft label matrix can be obtained. Based on the parameter information representing the second user label in the target user soft label matrix, the second user label can be determined.

[0096] It should be noted that soft tags can represent the probability of retaining a product tag that fits each user tag in the corresponding matrix, rather than mutual exclusion. This product tag belongs to only one user tag with a probability of 1.

[0097] According to embodiments of this disclosure, determining the target user soft tag matrix based on the probability transition matrix and the initial user soft tag matrix may include: obtaining the (i+1)th round user soft tag matrix based on the probability transition matrix and the i-th round user soft tag matrix, where the first round user soft tag matrix is ​​the initial user soft tag matrix, and i is an integer greater than or equal to 1. In response to determining that the information propagation probability between the first target product tag information and the second target product tag information obtained in the I-th round is less than a preset threshold, the I-th round user soft tag matrix is ​​determined as the target user soft tag matrix. The first target product tag information represents the product tag information corresponding to the node that obtained the second user tag information in the I-th round, and the second target product tag information is the product tag information that has a node association relationship with the first target product tag information.

[0098] It should be noted that the aforementioned first target product label may include multiple labels, and no limitation is made here.

[0099] Figure 5 A flowchart illustrating a training product relationship graph model according to an embodiment of the present disclosure is shown schematically.

[0100] like Figure 5 As shown, the method includes operations S510 to S540.

[0101] In operation S510, the probability transition matrix P and the initial user soft tag matrix F are obtained.

[0102] When operating S520, perform the propagation: F = PF.

[0103] During this process, the parameter information of the newly labeled user tags in F can be updated accordingly based on the calculation results.

[0104] In operation S530, determine whether F has converged. If yes, then execute operation S540; otherwise, execute operations S520 to S530.

[0105] In this process, whether F converges can be determined by judging whether the information propagation probability between the first target product label information and the second target product label information is less than a preset threshold. If it is determined that the information propagation probability between the first target product label information and the second target product label information is less than the preset threshold, then F can be determined to have converged.

[0106] In operation S540, a trained product relationship graph model is obtained based on the first product label, the second product label, and F.

[0107] Figure 6 A schematic diagram of a trained product relationship graph model according to an embodiment of the present disclosure is shown.

[0108] like Figure 6 As shown, the trained product relationship graph model includes nodes a, b, c, d, e, f, g, h, i, and edges ab, ac, bc, ad, ce, ef, e_i, eg, ef, fh, fi, gh, hi. Each node includes a product label and a user label.

[0109] It should be noted that for other user tags, such as risk tolerance tags and customer quality tags, the above method can be used to obtain the corresponding tag labeling results.

[0110] According to embodiments of this disclosure, determining the second user tag based on the target user soft tag matrix may also include: in response to determining that the user tags corresponding to the same second node information include multiple candidate user tags, determining the candidate user tag with the highest probability value among the multiple candidate user tags as the second user tag of the second node information.

[0111] For example, if the predicted user tags for a certain product tag include multiple user tags with probability values, the user tag with the highest probability value can be taken as the user tag corresponding to that product tag.

[0112] The above embodiments of this disclosure can further improve the accuracy of model output and effectively improve the accuracy of subsequent model-based recommended products.

[0113] According to embodiments of this disclosure, after obtaining a trained product recommendation model based on the above method, the model can be used to recommend products suitable for different users, with user tags as input.

[0114] Figure 7A flowchart illustrating a product recommendation method based on the trained product recommendation model described above, according to an embodiment of this disclosure, is shown.

[0115] like Figure 7 As shown, the method includes operations S710 to S720.

[0116] When operating S710, the target user tag is determined based on the target user attribute information of the target user.

[0117] When operating the S720, the target user tags are input into the trained product recommendation model to obtain the target products recommended to the target users.

[0118] According to embodiments of this disclosure, user tags matching the product targeted by each node in the trained product recommendation model can be determined. Based on this, when it is necessary to recommend products to a target user, the target product can be obtained by executing the above-described steps S710-S720.

[0119] For example, Table 7 shows a product recommendation result.

[0120] Table 7:

[0121]

[0122] According to embodiments of this disclosure, the target user tag may include at least a first user sub-tag and a second user sub-tag. The first user sub-tag has a first weight, and the second user sub-tag has a second weight, where the second weight is greater than the first weight. The process of inputting the target user tag into a product recommendation model to obtain target products recommended for the target user may include: in response to determining that there is no product among the products to be recommended that completely matches the target user tag, inputting the second user sub-tag into the product recommendation model to obtain a product that matches the second user sub-tag, which is then used as the target product. In response to determining that there is no product among the products to be recommended that matches the second user sub-tag, inputting the first user sub-tag into the product recommendation model to obtain a product that matches the first user sub-tag, which is then used as the target product.

[0123] For example, target user tags include customer group tags such as "adult users" and risk tolerance tags such as "low risk," with the risk tolerance tag carrying more weight than the customer group tag. If no user tag among the products to be recommended perfectly matches the target user, products with a "low risk tolerance" tag can be recommended first. If no products with a "low risk tolerance" tag exist among the products to be recommended, products with an "adult user" customer group tag can be recommended to the target user.

[0124] Through the above embodiments of this disclosure, a tag propagation algorithm is used to form community-based tags from numerous investment product types and a vast number of investment products, providing customers with accurate and comprehensive product profiles and recommendations, helping customers understand market investment trends, and achieving better investment guidance.

[0125] Based on the training method of the product recommendation model described above, this disclosure also provides a training device for the product recommendation model. The following will combine... Figure 8 The device is described in detail.

[0126] Figure 8 A schematic block diagram of a training apparatus for a product recommendation model according to an embodiment of the present disclosure is shown.

[0127] like Figure 8 As shown, the training device 800 for the product recommendation model in this embodiment includes a first acquisition module 810, a first determination module 820, and a first acquisition module 830.

[0128] The first acquisition module 810 is used to acquire a product relationship graph model constructed based on product tags. The product relationship graph model includes first node information, second node information, and edge information. The first node information includes a first product tag and a first user tag. The second node information only includes a second product tag. The edge information represents the association relationship between two nodes corresponding to the edge information. In one embodiment, the first acquisition module 810 can be used to perform the operation S210 described above, which will not be repeated here.

[0129] The first determining module 820 is used to determine the second user tag corresponding to the second product tag based on the first product tag, the first user tag, the second product tag, and the edge information. In one embodiment, the first determining module 820 can be used to perform the operation S220 described above, which will not be repeated here.

[0130] The first obtaining module 830 is used to obtain a trained product recommendation model based on the first node information, edge information, and second node information having a second product label and a second user label. In one embodiment, the first obtaining module 830 can be used to perform the operation S230 described above, which will not be repeated here.

[0131] According to embodiments of this disclosure, the first determining module includes a first determining unit and a second determining unit.

[0132] The first determining unit is used to determine the information propagation probability between the first node information and the second node information based on the second similarity between the first product label and the second product label.

[0133] The second determining unit is used to determine the second user tag based on the first user tag and the information propagation probability.

[0134] According to embodiments of this disclosure, the first determining unit includes a first determining subunit, a second determining subunit, and a third determining subunit.

[0135] The first determining subunit is used to determine the first edge weight information between the first node information and the second node information based on preset constraint parameters and the Euclidean distance between the first product label and the second product label.

[0136] The second determining subunit is used to determine the second edge weight information between every two nodes in the first node information and the second node information.

[0137] The third determining subunit is used to determine the information propagation probability based on the weight information of the first side and the weight information of the second side.

[0138] According to embodiments of this disclosure, the first determining module includes a third determining unit, a fourth determining unit, a fifth determining unit, and a sixth determining unit.

[0139] The third determining unit is used to determine the probability transition matrix based on the information propagation probability between every two nodes in the first node information and the second node information.

[0140] The fourth determining unit is used to determine the initial user soft tag matrix based on the first user tag and the initial parameter information representing the second user tag.

[0141] The fifth determining unit is used to determine the target user soft tag matrix based on the probability transition matrix and the initial user soft tag matrix, wherein the elements in the target user soft tag matrix include user tag information and probability values ​​corresponding to the user tag information.

[0142] The sixth determining unit is used to determine the second user tag based on the target user soft tag matrix.

[0143] According to embodiments of this disclosure, the fifth determining unit includes an obtaining subunit and a fourth determining subunit.

[0144] Obtain a sub-unit, which is used to obtain the user soft tag matrix for the (i+1)th round based on the probability transition matrix and the user soft tag matrix for the i-th round, where the user soft tag matrix for the 1st round is the initial user soft tag matrix, and i is an integer greater than or equal to 1.

[0145] The fourth determining subunit is used to determine the user soft tag matrix of the first round as the target user soft tag matrix in response to the determination that the information propagation probability between the first target product tag information and the second target product tag information obtained in the first round is less than a preset threshold. The first target product tag information represents the product tag information corresponding to the node that obtained the second user tag information in the first round, and the second target product tag information is the product tag information that has a node association relationship with the first target product tag information.

[0146] According to embodiments of this disclosure, the sixth determining unit includes the fifth determining subunit.

[0147] The fifth determining subunit is used to determine, in response to determining that the user label corresponding to the same second node information includes multiple candidate user labels, the candidate user label with the highest probability value among the multiple candidate user labels is determined as the second user label of the second node information.

[0148] According to embodiments of this disclosure, the training apparatus for the product recommendation model further includes a second acquisition module, a first conversion module, a second conversion module, a third acquisition module, a third conversion module, a second determination module, a third determination module, and a fourth determination module.

[0149] The second acquisition module is used to acquire the first product attribute information of the first product and the second product attribute information of the second product.

[0150] The first conversion module is used to perform structured conversion on the first product attribute information according to the first predefined mapping relationship to obtain the first product label. The first predefined mapping relationship represents the mapping relationship between various product attribute information and numerical information.

[0151] The second conversion module is used to perform structured conversion on the second product attribute information according to the first predefined mapping relationship to obtain the second product label.

[0152] The third acquisition module is used to acquire user attribute information of users related to the first product.

[0153] The third conversion module is used to perform structured conversion on user attribute information according to the second predefined mapping relationship to obtain the first user label. The second predefined mapping relationship represents the mapping relationship between various types of user attribute information and numerical information.

[0154] The second determining module is used to determine the first node information based on the first product tag and the first user tag corresponding to the same first product.

[0155] The third determining module is used to determine the second node information based on the second product label.

[0156] The fourth determining module is used to determine edge information based on the first similarity between every two product tags in the first product tag and the second product tag.

[0157] Based on the above product recommendation method, this disclosure also provides a product recommendation device. The following will be combined with... Figure 9 The device is described in detail.

[0158] Figure 9 A schematic block diagram of a product recommendation device according to an embodiment of the present disclosure is shown.

[0159] like Figure 9 As shown, the training device 800 for the product recommendation model in this embodiment includes a fifth determining module 910 and a second obtaining module 920.

[0160] The fifth determining module 910 is used to determine the target user tag based on the target user attribute information of the target user. In one embodiment, the fifth determining module 910 can be used to perform the operation S710 described above, which will not be repeated here.

[0161] The second obtaining module 920 is used to input the target user tags into the product recommendation model to obtain the target products recommended for the target user. The product recommendation model is a trained product recommendation model obtained according to the training method of this disclosure. In one embodiment, the second obtaining module 920 can be used to perform the operation S720 described above, which will not be repeated here.

[0162] According to embodiments of this disclosure, the target user tag includes at least a first user sub-tag and a second user sub-tag. The first user sub-tag has a first weight, and the second user sub-tag has a second weight, where the second weight is greater than the first weight. The second obtaining module includes a first obtaining unit and a second obtaining unit.

[0163] The first obtaining unit is used to, in response to determining that there is no product among the products to be recommended that completely matches the target user tag, input the second user sub-tag into the product recommendation model and obtain the product that matches the second user sub-tag as the target product.

[0164] The second obtaining unit is used to, in response to determining that there is no product in the products to be recommended that matches the second user sub-label, input the first user sub-label into the product recommendation model and obtain the product that matches the first user sub-label as the target product.

[0165] According to embodiments of this disclosure, any plurality of modules among the first acquisition module 810, the first determination module 820, and the first obtaining module 830, or the fifth determination module 910 and the second obtaining module 920, can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules can be combined with at least part of the functionality of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the first acquisition module 810, the first determination module 820, and the first obtaining module 830, or the fifth determination module 910 and the second obtaining module 920, can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these methods. Alternatively, at least one of the first acquisition module 810, the first determination module 820, and the first obtaining module 830, or the fifth determination module 910 and the second obtaining module 920, can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0166] Figure 10 A block diagram of an electronic device suitable for implementing at least one of the training methods and product recommendation methods of a product recommendation model, according to embodiments of the present disclosure, is illustrated.

[0167] like Figure 10 As shown, an electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage portion 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0168] RAM 1003 stores various programs and data required for the operation of electronic device 1000. Processor 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Processor 1001 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 1002 and / or RAM 1003. It should be noted that the programs may also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0169] According to embodiments of this disclosure, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to a bus 1004. The electronic device 1000 may also include one or more of the following components connected to the input / output (I / O) interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output (I / O) interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1010 as needed so that computer programs read from it can be installed into the storage section 1008 as needed.

[0170] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0171] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 1002 and / or RAM 1003 and / or one or more memories other than ROM 1002 and RAM 1003 described above.

[0172] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement the item recommendation method provided in the embodiments of this disclosure.

[0173] When the computer program is executed by the processor 1001, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0174] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 1009, and / or installed from a removable medium 1011. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0175] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by processor 1001, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0176] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0177] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0178] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0179] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A training method for a product recommendation model, comprising: Obtain a product relationship graph model constructed based on product tags, wherein the product relationship graph model includes first node information, second node information and edge information, the first node information includes a first product tag and a first user tag, the second node information includes only the second product tag, and the edge information represents the association relationship between two nodes corresponding to the edge information; Based on the second similarity between the first product label and the second product label, the information propagation probability between the first node information and the second node information is determined. Determine the probability transition matrix based on the information propagation probability between every two nodes in the first node information and the second node information; Based on the first user tag and the initial parameter information representing the second user tag, determine the initial user soft tag matrix; Based on the probability transition matrix and the initial user soft tag matrix, a target user soft tag matrix is ​​determined, wherein the elements in the target user soft tag matrix include user tag information and probability values ​​corresponding to the user tag information; Based on the target user soft tag matrix, determine the second user tag; and Based on the first node information, the edge information, and the second node information having the second product label and the second user label, a trained product recommendation model is obtained.

2. The method according to claim 1, wherein, The step of determining the information propagation probability between the first node information and the second node information based on the second similarity between the first product label and the second product label includes: Based on preset constraint parameters and the Euclidean distance between the first product label and the second product label, determine the first edge weight information between the first node information and the second node information; Determine the second edge weight information between every two nodes in the first node information and the second node information; and The information propagation probability is determined based on the first edge weight information and the second edge weight information.

3. The method according to claim 1, wherein, The step of determining the target user soft tag matrix based on the probability transition matrix and the initial user soft tag matrix includes: Based on the probability transition matrix and the user soft tag matrix of the i-th round, the user soft tag matrix of the (i+1)-th round is obtained, where the user soft tag matrix of the 1-th round is the initial user soft tag matrix, and i is an integer greater than or equal to 1; and In response to the determination that the information propagation probability between the first target product tag information and the second target product tag information obtained in the first round is less than a preset threshold, the user soft tag matrix of the first round is determined as the target user soft tag matrix, wherein the first target product tag information represents the product tag information corresponding to the node that obtained the second user tag information in the first round, and the second target product tag information is the product tag information that has a node association relationship with the first target product tag information.

4. The method according to claim 1, wherein, The step of determining the second user tag based on the target user soft tag matrix includes: In response to determining that the user tag corresponding to the same second node information includes multiple candidate user tags, the candidate user tag with the highest probability value among the multiple candidate user tags is determined as the second user tag of the second node information.

5. The method according to any one of claims 1-4, further comprising: Before obtaining the product relationship graph model constructed based on product tags. Obtain the first product attribute information of the first product and the second product attribute information of the second product; According to the first predefined mapping relationship, the first product attribute information is structurally transformed to obtain the first product tag, wherein the first predefined mapping relationship represents the mapping relationship between various types of product attribute information and numerical information; Based on the first predefined mapping relationship, the second product attribute information is structurally transformed to obtain the second product label; Obtain user attribute information of users related to the first product; According to the second predefined mapping relationship, the user attribute information is structurally transformed to obtain the first user tag, wherein the second predefined mapping relationship represents the mapping relationship between various types of user attribute information and numerical information; The first node information is determined based on the first product tag and the first user tag corresponding to the same first product; Based on the second product label, determine the second node information; and The edge information is determined based on the first similarity between every two product tags in the first product tag and the second product tag.

6. A product recommendation method, comprising: Determine target user tags based on the target user attribute information; as well as The target user tags are input into the product recommendation model to obtain the target products recommended for the target user, wherein the product recommendation model is a trained product recommendation model obtained by the method according to any one of claims 1-5.

7. The method according to claim 6, wherein, The target user tag includes at least a first user sub-tag and a second user sub-tag, the first user sub-tag has a first weight, the second user sub-tag has a second weight, and the second weight is greater than the first weight; The step of inputting the target user tags into the product recommendation model to obtain the target products recommended for the target user includes: In response to determining that there is no product among the products to be recommended that completely matches the target user tag, the second user sub-tag is input into the product recommendation model to obtain the product that matches the second user sub-tag, which is then used as the target product. as well as In response to determining that there is no product matching the second user sub-tag among the products to be recommended, the first user sub-tag is input into the product recommendation model to obtain the product matching the first user sub-tag, which is then used as the target product.

8. A training device for a product recommendation model, comprising: The first acquisition module is used to acquire a product relationship graph model constructed based on product tags. The product relationship graph model includes first node information, second node information, and edge information. The first node information includes a first product tag and a first user tag. The second node information only includes a second product tag. The edge information represents the association relationship between two nodes corresponding to the edge information. A first determining module is configured to: determine the information propagation probability between the first node information and the second node information based on a second similarity between the first product label and the second product label; determine a probability transition matrix based on the information propagation probability between every two nodes in the first and second node information; determine an initial user soft label matrix based on the first user label and initial parameter information representing the second user label; determine a target user soft label matrix based on the probability transition matrix and the initial user soft label matrix, wherein the elements of the target user soft label matrix include user label information and probability values ​​corresponding to the user label information; and determine a second user label based on the target user soft label matrix; and... The first acquisition module is used to obtain a trained product recommendation model based on the first node information, the edge information, and the second node information having the second product label and the second user label.

9. A product recommendation device, comprising: The fifth determination module is used to determine target user tags based on the target user attribute information of the target user; as well as The second obtaining module is used to input the target user tag into the product recommendation model to obtain the target product recommended for the target user, wherein the product recommendation model is a trained product recommendation model obtained by the device according to claim 8.

10. An electronic device, comprising: One or more processors; Storage device for storing one or more programs. Wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the method according to any one of claims 1 to 7.

11. A computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

12. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images