Recommendation method and related equipment for financial products

By building a user's knowledge graph, calculating recommendation scores based on attribute characteristics and operation information, and pushing financial products related to target business topics, the problem of low recommendation accuracy in financial products in the existing technology is solved and higher recommendation accuracy is achieved.

CN114417007BActive Publication Date: 2024-12-03CCB FINTECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111572374.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-21
Publication Date
2024-12-03
Estimated Expiration
2041-12-21

AI Technical Summary

Technical Problem

When recommending financial products, the existing financial product recommendation system is difficult to accurately reflect the user's idle funds based on the asset amount and historical purchase records, resulting in low recommendation accuracy.

Method used

By building a user's knowledge graph, determine the connection relationship between business topics and financial products, calculate the target recommendation score based on attribute characteristics, number of paths and user operation information, and push financial product information related to the target business topic.

Benefits of technology

It improves the accuracy and accuracy of financial product recommendations and can better meet users' investment needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114417007B_ABST
    Figure CN114417007B_ABST
Patent Text Reader

Abstract

The present invention provides a method for recommending financial products and related equipment, belonging to the technical field of recommendation systems, the method comprising: obtaining a knowledge graph corresponding to a first user to be recommended; determining target parameters corresponding to each business theme according to the knowledge graph; determining a target recommendation score corresponding to each business theme according to an initial recommendation score and target parameters corresponding to each business theme; determining a target business theme among various business themes according to the target recommendation score, and pushing product information of financial products associated with the target business theme to a terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score. In the present invention, the knowledge graph obtained through the financial products purchased or browsed by the user determines the target business theme that meets the user's expectations, thereby recommending the financial products associated with the target business theme to the user, thereby improving the accuracy of the recommendation of financial products, that is, the recommendation accuracy of financial products is high.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of recommendation systems, and in particular to a financial product recommendation method and related equipment. Background Art

[0002] As people's living standards improve, they have more and more idle funds. People can use their idle funds to purchase financial products and increase their wealth.

[0003] Currently, when customers purchase financial products, the financial product platform will make recommendations based on the customer’s asset amount and historical purchase records. However, the asset amount may be a small part of the user’s idle funds, and the recommendations for financial products are not accurate. Summary of the invention

[0004] The present invention provides a financial product recommendation method and related equipment to solve the problem of low recommendation accuracy of financial products.

[0005] In one aspect, the present invention provides a method for recommending financial products, comprising:

[0006] Acquire a knowledge graph corresponding to the first user to be recommended, wherein the product knowledge graph includes a connection relationship between multiple business topics and financial products, each of the business topics is connected to one or more financial products, and the business topics and the financial products are connected through common attribute features, and the financial products are financial products purchased or browsed by the first user;

[0007] Determine a target parameter corresponding to each of the business topics according to the knowledge graph, the target parameter including at least one of a first number of attribute features, a second number of paths, and operation information of a first user, the path being a connection path between the financial product and the business topic;

[0008] Determine a target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter;

[0009] A target business theme is determined from among the business themes according to the target recommendation score, and product information of financial products associated with the target business theme is pushed to a terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score.

[0010] In one embodiment, the target parameter includes the first quantity, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes:

[0011] determining a maximum number among each of said first numbers;

[0012] Determine a first correction value corresponding to each of the business topics according to the maximum number and a first number corresponding to each of the business topics, wherein the first number and the first correction value have a positive correlation;

[0013] According to the first correction value corresponding to the business theme, the initial recommendation score corresponding to the business theme is corrected to obtain the target recommendation score corresponding to the business theme.

[0014] In one embodiment, the target parameter includes the second number, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes:

[0015] Determine a path weight of each of the paths corresponding to the business subject according to the second number corresponding to the business subject, wherein the path weight is negatively correlated with the second number;

[0016] Determine the length of each path, and determine the product of the length of the path and the path weight to obtain a value corresponding to the path;

[0017] Determine a second correction value corresponding to the business theme according to the sum of the values ​​corresponding to each path corresponding to the business theme;

[0018] The initial recommendation score corresponding to the business theme is corrected according to the second correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0019] In one embodiment, the target parameter includes operation information of the first user, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes:

[0020] Acquire the operation behavior of the first user on each of the paths according to the operation information;

[0021] Determining a behavior weight corresponding to the path according to the type of operation behavior corresponding to the path;

[0022] Determining a third correction value corresponding to the business subject according to the sum of the behavior weights corresponding to each of the paths of the business subject;

[0023] The initial recommendation score corresponding to the business theme is corrected according to the third correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0024] In one embodiment, the target parameter includes operation information of the first user, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes:

[0025] Acquire operation information of each second user, wherein the first user is any of the second users;

[0026] Determine the number of user operations for each of the attribute features corresponding to the business topic according to the operation information corresponding to each of the second users;

[0027] Determine the popularity value of the business topic according to the number of user operations on each of the attribute features corresponding to the business topic;

[0028] Determine, according to the popularity value corresponding to the subject business, a fourth correction value corresponding to the subject business, wherein the popularity value and the fourth correction value are positively correlated;

[0029] The initial recommendation score corresponding to the business theme is corrected according to the fourth correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0030] In one embodiment, the step of obtaining the knowledge graph corresponding to the first user to be recommended includes:

[0031] Acquire operation information of the first user, and determine each financial product purchased or browsed by the first user according to the operation information of the first user;

[0032] Generate each business theme according to the attribute characteristics of each financial product, and determine the attribute characteristics matched by each business theme among the attribute characteristics;

[0033] Generate icons corresponding to the business theme, the attribute characteristics, and the financial product respectively;

[0034] The icon corresponding to the attribute feature is connected with the icon of the business subject matched by the attribute feature, and the icon corresponding to the attribute feature is connected with the icon of the financial product to which the attribute feature belongs, so as to obtain the knowledge graph.

[0035] In another aspect, the present invention further provides a financial product recommendation device, comprising:

[0036] an acquisition module, configured to acquire a knowledge graph corresponding to a first user to be recommended, wherein the product knowledge graph includes a connection relationship between a plurality of business topics and financial products, each of the business topics is connected to one or more financial products, and the business topics and the financial products are connected through a common attribute feature, and the financial products are financial products purchased or browsed by the first user;

[0037] a determination module, configured to determine a target parameter corresponding to each of the business topics according to the knowledge graph, wherein the target parameter includes at least one of a first number of attribute features, a second number of paths, and operation information of a first user, wherein the path is a connection path between the financial product and the business topic;

[0038] The determination module is further used to determine the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter;

[0039] The determination module is further used to determine a target business theme among the various business themes according to the target recommendation score, and push product information of financial products associated with the target business theme to a terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score.

[0040] On the other hand, the present invention also provides a financial product recommendation device, including: a memory and a processor;

[0041] The memory stores computer-executable instructions;

[0042] The processor executes the computer-executable instructions stored in the memory, so that the processor performs the financial product recommendation method as described above.

[0043] On the other hand, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the financial product recommendation method as described above.

[0044] On the other hand, the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the financial product recommendation method as described above.

[0045] The financial product recommendation method and related equipment provided by the present invention obtain the knowledge graph corresponding to the user to be recommended, and determine the target parameters such as the number of attribute features, the number of paths and the user's operation information corresponding to each business theme according to the knowledge graph, and then determine the target recommendation score corresponding to each business theme according to the initial recommendation score and the target parameters corresponding to each business theme, and determine the target business theme according to the target recommendation score, so as to push the product information of the financial product associated with the target business theme to the terminal associated with the user. In the present invention, the knowledge graph obtained through the financial products purchased or browsed by the user determines the target business theme that meets the user's expectations, so as to recommend the financial product associated with the target business theme to the user, thereby improving the accuracy of the recommendation of financial products, that is, the recommendation accuracy of financial products is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0047] Figure 1 A schematic diagram of a scenario for implementing a method for recommending financial products according to the present invention;

[0048] Figure 2 A schematic diagram of a flow chart of a first embodiment of a method for recommending financial products of the present invention;

[0049] Figure 3 This is a detailed flow chart of step S30 in the second embodiment of the method for recommending financial products of the present invention;

[0050] Figure 4 This is a detailed flow chart of step S30 in the third embodiment of the method for recommending financial products of the present invention;

[0051] Figure 5 This is a detailed flow chart of step S30 in the fourth embodiment of the method for recommending financial products of the present invention;

[0052] Figure 6 This is a detailed flow chart of step S30 in the fifth embodiment of the method for recommending financial products of the present invention;

[0053] Figure 7 A schematic diagram of a module of a financial product recommendation device of the present invention;

[0054] Figure 8 This is a schematic diagram of the hardware structure of the recommended device for the financial product of the present invention.

[0055] The above drawings show clear embodiments of the present disclosure, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present disclosure to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0056] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0057] It should be noted that the collection, storage, use, processing, transmission, provision and disclosure of information such as financial data or user data involved in the technical solution of this application are in compliance with the relevant laws and regulations and do not violate public order and good morals.

[0058] The present invention provides a method for recommending financial products. Figure 1 The scene diagram shown is implemented as shown. Figure 1 As shown, the financial product recommendation device 100 stores a user knowledge graph 110. The knowledge graph 110 includes multiple business topics, which are determined according to the characteristics of financial products. For example, two business topics, "stable income" and "high return rate", are generated according to financial products 1, 2, 3 and 4. Financial products are financial products that users browse or purchase. Business topics can be generated based on financial products, for example, Figure 1 Each financial product in generates two business themes, "stable income" and "high return rate", based on the generation paths a, b, c, and d. Business themes are connected with multiple attribute features. For example, "stable income" is connected with eight attribute features, namely, "T+1 arrival", "net value type", "R2 risk", "R1 risk", "currency type", "real-time arrival", ">N", and ">P". The attribute features are the attribute features of financial products. In addition, the knowledge graph 110 also includes the user's operation information ( Figure 1 The operation information includes viewing, sharing, following, placing orders, and purchasing. The financial product recommendation device 100 can determine the recommendation score of each business theme and the user according to the knowledge graph, so as to push the business theme with a higher recommendation score to the user's terminal.

[0059] The following specific embodiments are used to describe in detail the technical solutions of the present invention and how the technical solutions of the present application solve the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0060] Reference Figure 2 , Figure 2 This is the first embodiment of the method for recommending financial products of the present invention. The method for recommending financial products includes the following steps:

[0061] Step S10, obtaining the knowledge graph corresponding to the first user to be recommended, wherein the product knowledge graph includes a connection relationship between multiple business topics and financial products, each business topic is connected to one or more financial products, and the business topics and financial products are connected through common attribute features, and the financial products are financial products purchased or browsed by the first user.

[0062] In this embodiment, the execution subject is a financial product recommendation device. For ease of description, the device is used below to refer to the financial product recommendation device. The device stores a knowledge graph of users to be recommended. The user to be recommended is defined as the first user. The knowledge graph includes a connection relationship between multiple business topics and financial products, and each business topic is connected to one or more financial products, which are financial products purchased or browsed by the first user. Business topics and financial products are connected through common attribute features. Figure 1 As shown, financial product 1 is connected to the business theme "stable income" through the attribute features "T+1 arrival" and "net value type". Business themes are generated by extracting the attribute features of different financial products. For example, the number of reinvestors greater than 1,000 and the reinvestment ratio greater than 10% are defined as the business theme of "high return rate", and the maximum drawdown in the past year is less than 1% and the maximum drawdown in the past three years is less than 2% as the business theme of "low drawdown". The knowledge graph can be transmitted from an external device to the device, or it can be generated by the device for each financial product based on the operation information of the first user. The operation information includes information about financial products browsed and purchased by users. The operation information includes information such as the first user's viewing, sharing, attention, pending orders, and purchases of financial products.

[0063] Step S20, determining the target parameters corresponding to each business theme according to the knowledge graph, the target parameters including a first number of attribute features, a second number of paths, and at least one of the operation information of the first user, and the path is a connection path between the financial product and the business theme.

[0064] After determining the knowledge graph, the device can determine the target parameters corresponding to each business topic, and the target parameters include at least one of the first number of attribute features, the second number of paths, and the operation information of the first user. The first number is the number of attribute features connected to the business topic. The path is the connection path between the financial product and the business topic. Figure 1 Financial product 1 connects the two arrows with the business theme of “stable returns” through the characteristic attribute “T+1 arrival” to form the path of “stable returns”.

[0065] Step S30, determining a target recommendation score corresponding to each business theme according to the initial recommendation score corresponding to each business theme and the target parameter.

[0066] Each business theme has a corresponding initial recommendation score. The initial recommendation score w0 can be flexibly configured and intervened according to market conditions and business experience, and w0 can take a value between 0 and 1. That is, the initial recommendation score can be set by the customer manager of the financial product institution, and the set initial recommendation score is stored in the device. The device obtains the initial recommendation score corresponding to each business theme, and corrects the initial recommendation score based on the target parameter to determine the target recommendation score corresponding to the business theme. For example, the target parameter includes a first number of attribute features corresponding to the business theme. If the number of attribute features connected to the business theme is greater, the business theme is more in line with the first user's demand for purchasing financial products. Therefore, the correction value can be determined by the first number. The larger the first number, the larger the correction value, so that the target recommendation score obtained by correcting the initial recommendation score based on the correction value is larger.

[0067] Step S40, determining a target business theme among various business themes according to the target recommendation score, and pushing product information of financial products associated with the target business theme to a terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score.

[0068] After determining the target recommendation score corresponding to each business theme, it is necessary to determine whether the first user has a purchase intention for the financial product associated with the business theme based on the size of the target recommendation score. The device stores a preset score. If the target recommendation score is greater than the preset score, it can be determined that the business theme corresponding to the target recommendation score may be a business theme that the user is interested in. In this regard, the device determines the business theme with a target recommendation score greater than the preset score as the target business theme, and then pushes the product information of the financial product associated with the target business theme to the terminal associated with the first user.

[0069] In the technical solution provided in this embodiment, the knowledge graph corresponding to the user to be recommended is obtained, and the target parameters such as the number of attribute features, the number of paths, and the user's operation information corresponding to each business theme are determined based on the knowledge graph, and then the target recommendation score corresponding to each business theme is determined based on the initial recommendation score corresponding to each business theme and the target parameters, and the target business theme is determined based on the target recommendation score, so that the product information of the financial product associated with the target business theme is pushed to the terminal associated with the user. In the present invention, the knowledge graph obtained through the financial products purchased or browsed by the user determines the target business theme that meets the user's expectations, so that the financial products associated with the target business theme are recommended to the user, which improves the accuracy of the recommendation of financial products, that is, the recommendation accuracy of financial products is high.

[0070] Reference Figure 3 , Figure 3 This is the second embodiment of the method for recommending financial products of the present invention. Based on the first embodiment, step 30 includes:

[0071] Step S301, determining a maximum number among various first numbers.

[0072] In this embodiment, the target parameter includes a first quantity. The device determines a correction value based on the first quantity, and corrects the initial recommendation score based on the correction value to obtain a target recommendation score, and the correction value is defined as a first correction value. Specifically, the device determines the maximum number among each first quantity, and the maximum number represents the business topic with the most connected attribute characteristics.

[0073] Step S302: determining a first correction value corresponding to each business theme according to the maximum number and the first number corresponding to each business theme, wherein the first number and the first correction value are positively correlated.

[0074] The device determines the first correction value corresponding to each business theme based on the maximum number and the first number corresponding to each business theme. The more attribute features connected to the business theme, the larger the first correction value corresponding to the business theme, that is, the first number and the first correction value are positively correlated.

[0075] Step S303: correct the initial recommendation score corresponding to the business theme according to the first correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0076] After determining the first correction value corresponding to each business theme, it can be determined that the initial recommendation score of the business theme is corrected based on the first correction value of the business theme to obtain the target recommendation score corresponding to the business theme.

[0077] Specifically, the target recommendation score obtained by correcting the initial recommendation score with the first correction value can be represented by the following formula:

[0078] W = W0X(1-log 10 t num / log 10 t max_num )

[0079] Among them, w0 is the initial recommendation score, W is the target recommendation score, (1-log 10 t num / log 10 t max_num ) is the first correction value, t num is the first quantity, t max_num is the maximum number.

[0080] In the technical solution provided in this embodiment, the device determines the maximum number among each first number, and determines the first correction value corresponding to each business theme based on the maximum number and the first number corresponding to each business theme, and then corrects the initial recommendation score corresponding to the business theme based on the first correction value of the business theme, so as to accurately obtain the target recommendation score corresponding to the business theme.

[0081] Reference Figure 4 , Figure 4 This is the third embodiment of the method for recommending financial products of the present invention. Based on the first embodiment, step S30 includes:

[0082] Step S304: determining a path weight of each path corresponding to the business theme according to the second number corresponding to the business theme, wherein the path weight is negatively correlated with the second number.

[0083] In this embodiment, the target parameter includes the second number of paths connected to the business topic. The knowledge graph includes multiple paths, and each business topic is connected to multiple paths, for example, Figure 1 If the “stable return” in the example connects 6 attribute features, then “stable return” has six paths.

[0084] The device determines the path weight of each path based on the second number of paths connected to the business topic, and the larger the second number, the larger the path weight, that is, the path weight is negatively correlated with the second number. The path weight can be determined by Characterization, p_num j is the second number of paths included in the business topic numbered j. For example, if “robust income” has six paths, the path weight of each path of “robust income” is 1 / 6.

[0085] Step S305, determining the length of each path, and determining the product of the length of the path and the path weight to obtain a value corresponding to the path.

[0086] Each path has a corresponding length. For example, Figure 1 If "stable income" is connected to "currency type", the length of the path is 2; if financial products are connected to "T+1 arrival", and "T+1 arrival" is connected to "stable income", the length of the path is 3. The device determines the length of each path, and obtains the value corresponding to the path by multiplying the length of the determined path by the path weight.

[0087] Step S306: Determine a second correction value corresponding to the business theme according to the sum of the values ​​corresponding to each path corresponding to the business theme.

[0088] The device can determine the second correction value corresponding to the business theme based on the sum of the values ​​corresponding to each path of the business theme. It can be understood that the fewer the paths of the business theme, the greater the path weight, the greater the second correction value of the business theme, and thus the greater the target recommendation score.

[0089] Step S307: correct the initial recommendation score corresponding to the business theme according to the second correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0090] The device can correct the initial recommendation score based on the first correction value and the second correction value to obtain the target recommendation score. Specifically, the target recommendation score can be represented by the following formula:

[0091]

[0092] in, It is the sum of the values ​​corresponding to each path connected by the business topic.

[0093] In the technical solution provided in this embodiment, the second number of the device business theme paths determines the second correction value corresponding to the business theme, so that the initial recommendation score is corrected according to the second correction value to obtain an accurate target recommendation score.

[0094] Reference Figure 5 , Figure 5 This is the fourth embodiment of the method for recommending financial products of the present invention. Based on the first embodiment, step S30 includes:

[0095] Step S308: Acquire the operation behavior of the first user on each path according to the operation information.

[0096] In this embodiment, the target parameter includes the operation information of the first user. The device obtains the operation behavior of the first user on each path according to the operation information. The operation behavior includes weights, viewing, sharing, following, and placing orders. The device can configure corresponding behavior weights for different operation behaviors. In this regard, the device determines the operation behavior of the first user on the financial product based on the operation information, and each attribute feature entering the product has a corresponding operation behavior, so the operation behavior of the attribute feature connected to the business subject is the operation behavior of the path where the attribute feature is located.

[0097] Step S309: determining the behavior weight corresponding to the path according to the type of the operation behavior corresponding to the path.

[0098] Each operation behavior may be set with a corresponding behavior weight, and thus the device may determine the behavior weight corresponding to the path based on the type of the operation behavior of the path.

[0099] Step S310: determining a third correction value corresponding to the business theme according to the sum of the behavior weights corresponding to each path of the business theme.

[0100] Step S311, correcting the initial recommendation score corresponding to the business theme according to the third correction value corresponding to the business theme, to obtain the target recommendation score corresponding to the business theme.

[0101] The device obtains the behavior weight corresponding to each path of the business theme, thereby determining the sum of the behavior weights of each path of the business theme as the third correction value of the business theme, and then corrects the initial recommendation score according to the third correction value to obtain the target recommendation score.

[0102] In this embodiment, the device can correct the initial recommendation score by the first correction value, the second correction value, and the third correction value to obtain a target recommendation score, and the target recommendation score can be represented by the following formula:

[0103]

[0104] Among them, b is the behavior weight of the operation behavior of the path, It is the sum of the behavior weights of each path of the business topic.

[0105] In the technical solution provided in this embodiment, the device obtains the operation behavior of the first user on each path based on the user's operation information, determines the behavior weight corresponding to the path according to the type of the operation behavior corresponding to the path, and determines the third correction value according to the sum of the behavior weights of each path of the business theme. Finally, the third correction value of the business theme accurately corrects the initial recommendation score to obtain the target recommendation score of the business theme.

[0106] Reference Figure 6 , Figure 6This is the fifth embodiment of the method for recommending financial products of the present invention. Based on the first embodiment, step S30 includes:

[0107] Step S312: Acquire operation information of each second user, wherein the first user is any second user.

[0108] In this embodiment, the target parameter includes the operation information of the first user. The device needs to determine the correction value based on the popularity of the business topic. Specifically, the device obtains the operation information of each second user, and the second user is the user corresponding to all accounts stored in the device. The first user is any second user.

[0109] Step S313: Determine the number of user operations for each attribute feature corresponding to the business theme according to the operation information corresponding to each second user.

[0110] Step S314, determining the popularity value of the business topic according to the number of user operations on each attribute feature corresponding to the business topic.

[0111] The operation information is the operation information of the second user on the attribute characteristics of the second financial product. Therefore, the device can determine the number of user operations on each attribute characteristic of each business theme according to the operation information of each second user. For example, according to each operation information, it is determined that the financial product has been operated by 100 users, then the number of user operations on the attribute characteristics corresponding to the financial product can also be 100 times, that is, the number of user operations on the attribute characteristics of the financial product connected to the business theme is also 100 times. The business theme includes multiple attribute characteristics, each of which has a corresponding number of user operations. The device can calculate the total number of user operations on the business theme based on the number of user operations on each attribute characteristic of the business theme. The heat value of the business theme can be determined by the total number of user operations. The heat value can be normalized, that is, the heat value is between 0-1.

[0112] Step S315, determining a fourth correction value corresponding to the business theme according to the popularity value corresponding to the theme business, and the popularity value and the fourth correction value are positively correlated.

[0113] Step S316, correcting the initial recommendation score corresponding to the business theme according to the fourth correction value corresponding to the business theme, to obtain the target recommendation score corresponding to the business theme.

[0114] The device can determine the fourth correction value of the business theme based on the popularity value, and the larger the popularity value, the larger the fourth correction value. The device corrects the initial recommendation score corresponding to the business theme based on the fourth correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0115] In this embodiment, the device can correct the initial recommendation score by the first correction value, the second correction value, the third correction value, and the fourth correction value to obtain a target recommendation score, and the target recommendation score can be represented by the following formula:

[0116] W=w0×h t

[0117] Among them, h t That is the fourth correction value determined by the heat value.

[0118] It should be noted that the device can obtain a target correction value based on the combination operation of each of the second correction values ​​mentioned above and the first correction value, and then correct the initial recommendation score by the target correction value to obtain the target recommendation score. For example, the target recommendation score can be represented by the following formula:

[0119]

[0120] In the technical solution provided in this embodiment, the device determines the popularity value of the business theme through the operation information of all users, and thus determines the fourth correction value of the business theme according to the popularity value. Finally, the fourth correction value of the business theme accurately corrects the initial recommendation score to obtain the target recommendation score of the business theme.

[0121] In one embodiment, step S10 includes:

[0122] Obtain the operation information of the first user, and determine the various financial products purchased or browsed by the first user based on the operation information of the first user

[0123] Generate various business themes according to the attribute characteristics of various financial products, and determine the attribute characteristics matched by each business theme among various attribute characteristics;

[0124] Generate icons corresponding to business themes, attribute features, and financial products respectively;

[0125] The icon corresponding to the attribute feature is connected with the icon of the business theme matched by the attribute feature, and the icon corresponding to the attribute feature is connected with the icon of the financial product to which the attribute feature belongs, so as to obtain a knowledge graph.

[0126] In this embodiment, the device can generate a knowledge graph. Specifically, the device obtains the operation information of the first user, and then determines the financial products purchased or browsed by the first user according to the operation information of the first user. The device generates each business theme according to each first financial product. The device then determines the attribute characteristics matched by each business theme. For example, referring to Figure 1 "Stable returns" matches the six attribute characteristics of "T+1 arrival", "net value type", "R2 risk", "R1 risk", "currency type" and "real-time arrival".

[0127] The device generates icons for business themes, attribute features, and financial products respectively, and then connects the icons corresponding to the attribute features with the icons of the business themes that match the attribute features, and connects the icons corresponding to the attribute features with the icons of the financial products to which the attribute features belong, to obtain a knowledge graph.

[0128] In the technical solution provided in this embodiment, the device can accurately generate the knowledge graph of the first user based on the operation information of the first user.

[0129] The present invention also provides a financial product recommendation device, referring to Figure 7 , the financial product recommendation device 700 includes:

[0130] The acquisition module 710 is used to acquire the knowledge graph corresponding to the first user to be recommended, wherein the product knowledge graph includes a connection relationship between multiple business topics and financial products, each business topic is connected to one or more financial products, and the business topic and the financial product are connected through a common attribute feature, and the financial product is a financial product purchased or browsed by the first user;

[0131] A determination module 720, configured to determine a target parameter corresponding to each business theme according to the knowledge graph, where the target parameter includes at least one of a first number of attribute features, a second number of paths, and operation information of the first user, where the path is a connection path between the financial product and the business theme;

[0132] A determination module 720, configured to determine a target recommendation score corresponding to each business theme according to the initial recommendation score corresponding to each business theme and the target parameter;

[0133] The determination module 720 is used to determine a target business theme among various business themes according to the target recommendation score, and push product information of financial products associated with the target business theme to the terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score.

[0134] In one embodiment, the financial product recommendation device 700 includes:

[0135] A determination module 720, configured to determine a maximum number among the first numbers;

[0136] A determination module 720, configured to determine a first correction value corresponding to each business theme according to the maximum number and the first number corresponding to each business theme, wherein the first number and the first correction value are positively correlated;

[0137] The correction module is used to correct the initial recommendation score corresponding to the business theme according to the first correction value corresponding to the business theme, so as to obtain the target recommendation score corresponding to the business theme.

[0138] In one embodiment, the financial product recommendation device 700 includes:

[0139] A determination module 720, configured to determine a path weight of each path corresponding to the business theme according to the second number corresponding to the business theme, wherein the path weight is negatively correlated with the second number;

[0140] A determination module 720 is used to determine the length of each path, and determine the product of the length of the path and the path weight to obtain a value corresponding to the path;

[0141] A determination module 720, configured to determine a second correction value corresponding to the business theme according to the sum of the values ​​corresponding to the paths corresponding to the business theme;

[0142] The correction module is used to correct the initial recommendation score corresponding to the business theme according to the second correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0143] In one embodiment, the financial product recommendation device 700 includes:

[0144] An acquisition module 710 is used to acquire the operation behavior of the first user on each path according to the operation information;

[0145] A determination module 720, configured to determine a behavior weight corresponding to the path according to a type of operation behavior corresponding to the path;

[0146] A determination module 720, configured to determine a third correction value corresponding to the business theme according to the sum of the behavior weights corresponding to each path of the business theme;

[0147] The correction module is used to correct the initial recommendation score corresponding to the business theme according to the third correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

[0148] In one embodiment, the financial product recommendation device 700 includes:

[0149] An acquisition module 710 is used to acquire operation information of each second user, wherein the first user is any second user;

[0150] A determination module 720, configured to determine the number of user operations for each attribute feature corresponding to the business theme according to the operation information corresponding to each second user;

[0151] A determination module 720, configured to determine a popularity value of a business topic according to the number of user operations on each attribute feature corresponding to the business topic;

[0152] A determination module 720 is used to determine a fourth correction value corresponding to the business theme according to the popularity value corresponding to the theme business, and the popularity value and the fourth correction value are positively correlated;

[0153] The correction module is used to correct the initial recommendation score corresponding to the business theme according to the fourth correction value corresponding to the business theme, so as to obtain the target recommendation score corresponding to the business theme.

[0154] In one embodiment, the financial product recommendation device 700 includes:

[0155] The acquisition module 710 is used to acquire the operation information of the first user, and determine each financial product purchased or browsed by the first user according to the operation information of the first user;

[0156] A generation module, used to generate various business themes according to the attribute characteristics of various financial products, and determine the attribute characteristics matched by each business theme among the various attribute characteristics;

[0157] A generation module, used to generate icons corresponding to business themes, attribute features and financial products respectively;

[0158] The connection module is used to connect the icon corresponding to the attribute feature with the icon of the business theme matched by the attribute feature, and to connect the icon corresponding to the attribute feature with the icon of the financial product to which the attribute feature belongs, so as to obtain a knowledge graph.

[0159] Figure 8 The figure is a hardware structure diagram of a financial product recommendation device according to an exemplary embodiment.

[0160] The financial product recommendation device 800 may include: a processor 801, such as a CPU, a memory 802, and a transceiver 803. Those skilled in the art will appreciate that Figure 8 The structure shown in does not constitute a limitation on the recommended device of the financial product, and may include more or less components than shown, or combine certain components, or arrange the components differently. The memory 802 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0161] The processor 801 may call the computer program stored in the memory 802 to complete all or part of the steps of the above-mentioned financial product recommendation method.

[0162] The transceiver 803 is used to receive information sent by an external device and send information to the external device.

[0163] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a financial product recommendation device, enables the financial product recommendation device to perform the above-mentioned financial product recommendation method.

[0164] A computer program product includes a computer program. When the computer program is executed by a processor of a financial product recommendation device, the financial product recommendation device is enabled to perform the above-mentioned financial product recommendation method.

[0165] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be regarded as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0166] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for recommending financial products, characterized in that: include: Acquire a knowledge graph corresponding to the first user to be recommended, wherein the knowledge graph includes a connection relationship between multiple business topics and financial products, each of the business topics is connected to one or more financial products, and the business topics and the financial products are connected through common attribute features, and the financial products are financial products purchased or browsed by the first user; wherein the business topics are determined according to the characteristics of the financial products; and the attribute features are attribute features of the financial products; Determine a target parameter corresponding to each of the business topics according to the knowledge graph, the target parameter including at least one of a first number of attribute features, a second number of paths, and operation information of a first user, the path being a connection path between the financial product and the business topic; Determine a target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter; Determining a target business theme from among the business themes according to the target recommendation score, and pushing product information of financial products associated with the target business theme to a terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score; The target parameter includes the first quantity and operation information of the first user, the operation information includes at least one of viewing, sharing, following, placing an order, and purchasing, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes: determining a maximum number among each of said first numbers; Determine a first correction value corresponding to each of the business topics according to the maximum number and a first number corresponding to each of the business topics, wherein the first number and the first correction value have a positive correlation; Acquire the operation behavior of the first user on each of the paths according to the operation information; Determining a behavior weight corresponding to the path according to the type of operation behavior corresponding to the path; Determining a third correction value corresponding to the business subject according to the sum of the behavior weights corresponding to each of the paths of the business subject; The initial recommendation score corresponding to the business theme is corrected according to the first correction value and the third correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

2. The method for recommending financial products according to claim 1, characterized in that: The target parameter includes the second number and operation information of the first user, the operation information includes at least one of viewing, sharing, following, placing an order, and purchasing, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes: Determine a path weight of each of the paths corresponding to the business subject according to the second number corresponding to the business subject, wherein the path weight is negatively correlated with the second number; Determine the length of each path, and determine the product of the length of the path and the path weight to obtain a value corresponding to the path; Determine a second correction value corresponding to the business theme according to the sum of the values ​​corresponding to each path corresponding to the business theme; Acquire the operation behavior of the first user on each of the paths according to the operation information; Determining a behavior weight corresponding to the path according to the type of operation behavior corresponding to the path; Determining a third correction value corresponding to the business subject according to the sum of the behavior weights corresponding to each of the paths of the business subject; The initial recommendation score corresponding to the business theme is corrected according to the second correction value and the third correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

3. The method for recommending financial products according to claim 1, characterized in that: The target parameter includes operation information of the first user, and the step of determining the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter includes: Acquire operation information of each second user, wherein the first user is any of the second users; Determine the number of user operations for each of the attribute features corresponding to the business topic according to the operation information corresponding to each of the second users; Determine the popularity value of the business topic according to the number of user operations on each of the attribute features corresponding to the business topic; Determine, according to the popularity value corresponding to the subject business, a fourth correction value corresponding to the subject business, wherein the popularity value and the fourth correction value are positively correlated; The initial recommendation score corresponding to the business theme is corrected according to the fourth correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

4. The method for recommending financial products according to any one of claims 1 to 3, characterized in that: The step of obtaining the knowledge graph corresponding to the first user to be recommended includes: Acquire operation information of the first user, and determine each financial product purchased or browsed by the first user according to the operation information of the first user; Generate each business theme according to the attribute characteristics of each financial product, and determine the attribute characteristics matched by each business theme among the attribute characteristics; Generate icons corresponding to the business theme, the attribute characteristics, and the financial product respectively; The icon corresponding to the attribute feature is connected with the icon of the business subject matched by the attribute feature, and the icon corresponding to the attribute feature is connected with the icon of the financial product to which the attribute feature belongs, so as to obtain the knowledge graph.

5. A financial product recommendation device, characterized in that: include: an acquisition module, configured to acquire a knowledge graph corresponding to the first user to be recommended, wherein the knowledge graph includes a connection relationship between a plurality of business themes and financial products, each of the business themes is connected to one or more financial products, and the business themes and the financial products are connected through a common attribute feature, the financial products are financial products purchased or browsed by the first user; the business themes are determined according to the characteristics of the financial products; and the attribute features are the attribute features of the financial products; a determination module, configured to determine a target parameter corresponding to each of the business topics according to the knowledge graph, wherein the target parameter includes at least one of a first number of attribute features, a second number of paths, and operation information of a first user, wherein the path is a connection path between the financial product and the business topic; The determination module is further used to determine the target recommendation score corresponding to each of the business topics according to the initial recommendation score corresponding to each of the business topics and the target parameter; The determination module is further configured to determine a target business theme from among the business themes according to the target recommendation score, and push product information of financial products associated with the target business theme to a terminal associated with the first user, wherein the target recommendation score of the target business theme is greater than a preset score; The target parameters include the first quantity and operation information of the first user, the operation information includes at least one of viewing, sharing, following, placing an order, and purchasing, and the determination module is specifically used to determine the maximum quantity among each of the first quantities; determine the first correction value corresponding to each of the business themes according to the maximum quantity and the first quantity corresponding to each of the business themes, and the first quantity and the first correction value are positively correlated; obtain the operation behavior of the first user on each of the paths according to the operation information; determine the behavior weight corresponding to the path according to the type of operation behavior corresponding to the path; determine the third correction value corresponding to the business theme according to the sum of the behavior weights corresponding to each of the paths of the business theme; correct the initial recommendation score corresponding to the business theme according to the first correction value and the third correction value corresponding to the business theme to obtain the target recommendation score corresponding to the business theme.

6. A financial product recommendation device, characterized in that: include: Memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor executes the financial product recommendation method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for recommending financial products according to any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for recommending financial products according to any one of claims 1 to 4 is implemented.