CN116664241A

CN116664241A - Financial product recommendation method, device, equipment and medium

Info

Publication number: CN116664241A
Application number: CN202310661297.5A
Authority: CN
Inventors: 雷声渊; 徐李融; 孙文豪; 杨平
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2023-06-06
Filing date: 2023-06-06
Publication date: 2023-08-29

Abstract

The embodiment of the disclosure provides a financial product recommendation method which can be applied to the technical field of computers and the technical field of big data. The method comprises the following steps: and acquiring user login information and user history query information. And obtaining first associated data through data dimension reduction processing, and obtaining a preference value of a user on a product through a preference value generation model. A recommendation list is determined, the recommendation list including a plurality of recommended financial products. Product parameters and dynamic information corresponding to a plurality of recommended financial products in the recommendation list are acquired, and dynamic data of the financial products are generated. And outputting a plurality of recommended financial products and dynamic data in the recommendation list. The present disclosure also provides a financial product recommendation apparatus, a computing device, a medium, and a program product.

Description

Financial product recommendation method, device, equipment and medium

Technical Field

The present disclosure relates to the field of computer technology, and in particular, to a method, apparatus, device, medium, and program product for recommending a financial product.

Background

With the continuous development of internet technology, the financial products of banks are increased year by year along with customer demands and market demands, and marketing of various financial products of banks is also transferred from offline to online, and the financial products are recommended to users through terminal data analysis, so that labor cost is saved, and the probability that the recommended financial products are accepted by the users is improved.

At present, a financial product recommendation method applied in the market generally obtains and analyzes historical query data of a user, and matches a suitable financial product for the user to recommend according to an analysis result. Because of the large number of customers of banks, the financial product recommendation method needs to analyze a large amount of data, so that the calculated amount of a computer is increased, the cost is increased, and the efficiency is low. In addition, most bank clients do not have professional financial knowledge, so that the specific effect of recommended financial products is difficult to accurately understand, staff is also required to be inquired, the workload of the staff is increased, and the working efficiency is reduced.

Disclosure of Invention

In view of the foregoing, the present disclosure provides a financial product recommendation method, apparatus, device, medium, and program product.

According to a first aspect of the present disclosure, there is provided a financial product recommendation method, the method comprising:

acquiring user login information;

responding to the user login information and acquiring user history inquiry information;

based on the user history query information, obtaining first associated data through data dimension reduction processing;

obtaining a preference value of a user for a product through a preference value generation model according to the first associated data, wherein the preference value generation model is obtained through pre-training;

Determining a recommendation list according to the preference value, wherein the recommendation list comprises a plurality of recommended financial products;

based on the recommendation list, obtaining product parameters and dynamic information corresponding to a plurality of recommended financial products in the recommendation list, wherein the dynamic information comprises: financial product browsing amount, financial product attention amount, financial product transaction amount and financial product evaluation;

generating dynamic data of the financial product according to the product parameters and the dynamic information; and

and outputting a plurality of recommended financial products and dynamic data in the recommendation list.

According to an embodiment of the present disclosure, in response to user login information, obtaining the user history query information includes:

judging whether the historical query information of the user is empty or not;

if the historical query information of the user is empty, acquiring the login information of the historical user, and comparing the similarity between the login information of the user and the login information of the historical user to obtain the login information of the similar user; and

and acquiring the historical query information of the similar user according to the login information of the similar user, and taking the historical query information as the historical query information of the user.

According to an embodiment of the present disclosure, performing similarity comparison between the user login information and the historical user login information, and obtaining the login information of the similar user includes:

Acquiring first characteristic data of the user login information;

acquiring second characteristic data of the historical user login information; and

comparing the first characteristic data with the second characteristic data, and taking the historical user login information of which the second characteristic data is the same as the first characteristic data as the login information of a similar user.

According to an embodiment of the present disclosure, obtaining, through data processing, first associated data based on the user history query information includes:

labeling the second associated data in the user history query information based on the second characteristic data to obtain labeled third associated data;

performing data cleaning on the third associated data to obtain cleaned fourth associated data, wherein the data cleaning comprises removing abnormal data in the third associated data, and the abnormal data comprises incomplete data, error data and repeated data; and

and carrying out data feature fusion on the fourth associated data to obtain first associated data.

According to an embodiment of the present disclosure, the preference value generating model is a hidden-type relationship model, and pre-training the preference value generating model includes:

Acquiring a training data set, wherein the training data set comprises first association data for training;

acquiring implicit characteristics common to users and financial products in the training data set based on the first association data for training;

acquiring a first weight of a user on the implicit characteristic, and constructing a first weight matrix of the user on the implicit characteristic;

acquiring a second weight of the financial product to the implicit characteristic, and constructing a second weight matrix of the financial product to the implicit characteristic;

based on the implicit characteristics, fusing the first weight matrix and the second weight matrix to obtain a preference value matrix of a user on the financial product; and

and calculating the preference value of the user for the financial product according to the preference value matrix of the user for the financial product.

According to an embodiment of the present disclosure, pre-training the preference value generation model further includes:

acquiring a test data set, wherein the test data set comprises first association data for testing;

presetting a test data set verification success rate threshold;

verifying the data in the test data set, and obtaining a verification success rate; and

when the verification success rate is greater than the success rate threshold value, determining the hidden relation model,

Wherein verifying the data in the test dataset comprises:

obtaining a real preference value of a test user for a test product according to the first association data for the test;

inputting the test user into the hidden relation model to obtain a test preference value of a test product;

carrying out loss optimization on the real preference value and the test preference value in a mean square error mode to obtain a loss value;

presetting a loss threshold; and

if the loss value is within the loss threshold, the verification is successful.

According to a second aspect of the present disclosure, there is provided a financial product recommendation device comprising:

the first acquisition module is used for acquiring user login information;

the second acquisition module is used for responding to the user login information and acquiring the user history inquiry information;

the data processing module is used for obtaining first associated data through data dimension reduction processing based on the user history query information;

the model generation module is used for generating a model according to the first associated data through a preference value to obtain a preference value of a user for a product, wherein the preference value generation model is obtained through pre-training;

a first determining module, configured to determine a recommendation list according to the preference value, where the recommendation list includes a plurality of recommended financial products;

A third obtaining module, configured to obtain, based on the recommendation list, product parameters and dynamic information corresponding to a plurality of recommended financial products in the recommendation list, where the dynamic information includes: financial product browsing amount, financial product attention amount, financial product transaction amount and financial product evaluation;

the data generation module is used for generating dynamic data of the financial product according to the product parameters and the dynamic information; and

and the output module is used for outputting a plurality of recommended financial products and dynamic data in the recommendation list.

According to a third aspect of the present disclosure, there is provided an electronic device comprising:

one or more processors;

storage means for storing one or more programs,

wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the financial product recommendation method described above.

According to a fourth aspect of the present disclosure, a computer-readable storage medium has stored thereon executable instructions that, when executed by a processor, cause the processor to perform the above-described financial product recommendation method.

There is also provided in accordance with a fifth aspect of the present disclosure a computer program product comprising a computer program which, when executed by a processor, implements the above-described financial product recommendation method.

Drawings

The foregoing and other objects, features and advantages of the disclosure will be more apparent from the following description of embodiments of the disclosure with reference to the accompanying drawings, in which:

FIG. 1 schematically illustrates an application scenario diagram of a financial product recommendation method and apparatus according to an embodiment of the present disclosure;

FIG. 2 schematically illustrates a flow chart of a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 3 schematically illustrates a flowchart for obtaining historical user query information in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 4 schematically illustrates a flow chart of similarity comparison in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 5 schematically illustrates a flowchart of obtaining first associated data through data processing in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 6 schematically illustrates a flow chart of a training preference value generation model in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 7 schematically illustrates a flowchart of a verification preference value generation model in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 8 schematically illustrates a flow chart for validating data in a test dataset in a financial product recommendation method according to an embodiment of the disclosure;

FIG. 9 schematically illustrates a flowchart of a user querying a financial product in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 10 schematically illustrates a flow chart of information interaction when a user queries a financial product in a financial product recommendation method according to an embodiment of the present disclosure;

FIG. 11 schematically illustrates a block diagram of a financial product recommendation device, according to an embodiment of the present disclosure;

FIG. 12 schematically illustrates a block diagram of a second acquisition module in a financial product recommendation device, according to an embodiment of the present disclosure;

FIG. 13 schematically illustrates a block diagram of a first contrast module in a financial product recommendation device, according to an embodiment of the present disclosure;

FIG. 14 schematically illustrates a block diagram of a data processing module in a financial product recommendation device, in accordance with an embodiment of the present disclosure;

FIG. 15 schematically illustrates a block diagram of a model generation module of a financial product recommendation device, in accordance with an embodiment of the present disclosure;

fig. 16 schematically illustrates a block diagram of an electronic device adapted to implement a financial product recommendation method, in accordance with an embodiment of the present disclosure.

Detailed Description

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is only exemplary and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. In addition, in the following description, descriptions of well-known structures and techniques are omitted so as not to unnecessarily obscure the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and/or the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.

All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It should be noted that the terms used herein should be construed to have meanings consistent with the context of the present specification and should not be construed in an idealized or overly formal manner.

Where expressions like at least one of "A, B and C, etc. are used, the expressions should generally be interpreted in accordance with the meaning as commonly understood by those skilled in the art (e.g.," a system having at least one of A, B and C "shall include, but not be limited to, a system having a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.). Where a formulation similar to at least one of "A, B or C, etc." is used, in general such a formulation should be interpreted in accordance with the ordinary understanding of one skilled in the art (e.g. "a system with at least one of A, B or C" would include but not be limited to systems with a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).

Some of the block diagrams and/or flowchart illustrations are shown in the figures. It will be understood that some blocks of the block diagrams and/or flowchart illustrations, or combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable control apparatus, such that the instructions, when executed by the processor, create means for implementing the functions/acts specified in the block diagrams and/or flowchart.

First, technical terms appearing herein are explained as follows:

labeling: according to the requirements of business scenes, highly accurate feature labels are obtained by applying algorithms such as abstraction, induction, reasoning and the like to target objects (including static and dynamic features). The label consists of a label name and a label value, and is marked upwards on the target object.

The embodiment of the disclosure provides a financial product recommendation method, which comprises the following steps: user login information is acquired, and user history query information is acquired in response to the user login information. And obtaining first associated data through data dimension reduction processing based on the user history query information. And obtaining a preference value of the user for the product through a preference value generation model according to the first association data, wherein the preference value generation model is obtained through pre-training. And determining a recommendation list according to the preference value, wherein the recommendation list comprises a plurality of recommended financial products. Based on the recommendation list, obtaining product parameters and dynamic information corresponding to a plurality of recommended financial products in the recommendation list, wherein the dynamic information comprises: financial product browsing amount, financial product attention amount, financial product transaction amount and financial product evaluation. And generating dynamic data of the financial product according to the product parameters and the dynamic information. And outputting a plurality of recommended financial products and dynamic data in the recommendation list.

According to the method, the preference value of the user for the product is obtained through the construction model, the data is reduced in dimension, the preference value is generated to obtain the technical means of the recommendation list, the technical problems that complicated data are simplified and the product is accurately recommended to the user are solved, and the technical effects of reducing the calculated amount of a computer and saving the calculation resources are achieved. And through recommending the corresponding dynamic data of the product, can make users understand the income that the financial product brings to oneself more easily, the product suitable for oneself of the choice of being convenient for, do not need to ask the staff any more, has raised working efficiency.

Fig. 1 schematically illustrates an application scenario diagram of a financial product recommendation method and apparatus according to an embodiment of the present disclosure. It should be noted that fig. 1 is merely an example of a scenario in which embodiments of the present disclosure may be applied to assist those skilled in the art in understanding the technical content of the present disclosure, but does not mean that embodiments of the present disclosure may not be used in other devices, systems, environments, or scenarios.

As shown in fig. 1, the application scenario 100 according to this embodiment may include a plurality of application terminals and application servers. For example, the plurality of application terminals includes an application terminal 101, an application terminal 102, an application terminal 103, and the like. The network 104 is used as a medium to provide communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.

The user may interact with the application server 105 via the network 104 using the application terminal devices 101, 102, 103 to receive or send messages or the like. Various application programs such as shopping class applications, web browser applications, search class applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only) may be installed on the application terminal devices 101, 102, 103.

The terminal devices 101, 102, 103 may be a variety of electronic devices having a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop and desktop computers, and the like.

The server 105 may be a server providing various services, such as a background management server (by way of example only) providing support for websites browsed by users using the terminal devices 101, 102, 103. The background management server may analyze and process the received data such as the user request, and feed back the processing result (e.g., the web page, information, or data obtained or generated according to the user request) to the terminal device.

It should be noted that the financial product recommendation method provided by the embodiments of the present disclosure may be generally performed by the server 105. Accordingly, the financial product recommendation device provided by the embodiments of the present disclosure may be generally disposed in the server 105. The financial product recommendation method provided by the embodiments of the present disclosure may also be performed by a server or server cluster that is different from the server 105 and is capable of communicating with the terminal devices 101, 102, 103 and/or the server 105. Accordingly, the financial product recommendation device provided by the embodiments of the present disclosure may also be provided in a server or a server cluster that is different from the server 105 and is capable of communicating with the terminal devices 101, 102, 103 and/or the server 105.

It should be understood that the number of terminal devices, networks and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

The financial product recommendation method of the disclosed embodiment will be described in detail below with reference to fig. 2 to 10 based on the scenario described in fig. 1. It should be noted that the above application scenario is only shown for the convenience of understanding the spirit and principles of the present disclosure, and the embodiments of the present disclosure are not limited in any way in this respect. Rather, embodiments of the present disclosure may be applied to any scenario where applicable.

Fig. 2 schematically illustrates a flowchart of a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 2, the method 200 includes steps S201 to S208.

Step S201, user login information is acquired.

When a user logs in the system, the system can judge whether the user is a registered user or not, and if the user is a registered user, the user directly enters the system; if the user is an unregistered user, the registration system is jumped to enable the user to log in the system after registering.

The obtaining of the user login information comprises the following steps: information registered by a user in a server is acquired. For example, the information of user registration includes: the gender, age, occupation, hobbies, etc. of the user.

Step S202, responding to the user login information and acquiring user history inquiry information.

Fig. 3 schematically illustrates a flowchart for acquiring historical user query information in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 3, the method 300 includes steps S301 to S203.

Step S301, determining whether the user history query information is empty.

If the historical query information of the user is empty, the user is an old user but the product information is not queried, or the user is a newly registered user.

Step S302, if the user history inquiry information is empty, obtaining history user login information, and comparing the similarity between the user login information and the history user login information to obtain login information of a similar user.

Fig. 4 schematically illustrates a flowchart for similarity comparison in the financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 4, the method 400 includes steps S401 to S403.

Step S401, acquiring first feature data of the user login information.

For example, the age of the user is 25 years, the gender is male, the occupation is sales, the hobbies are running and basketball.

Step S402, second characteristic data of the historical user login information is obtained.

Second characteristic data of a plurality of historical user login information, such as age, gender, occupation, interests and the like, is acquired.

Step S403, comparing the first feature data with the second feature data, and using the historical user login information of the second feature data, which is the same as the first feature data, as the login information of the similar user.

For example, a user whose age is 25 years, sex is male, occupation is sales, interest is fishing, and basketball is found as a similar user among the history user login information, and the similar user's login information is taken as the login information of the target user.

Of course, the user with the age of 23 to 28 years, sex of male, occupation of sales or salesman, interest of sports can be found in the historical user login information, and the login information of the similar user is used as the login information of the target user.

By acquiring the login information of the user and the characteristics in the login information of the historical user, the user is matched with similar users, and the accuracy of recommended products is improved conveniently.

Referring back to fig. 3, in step S303, history query information of the similar user is obtained as the history query information of the user according to the login information of the similar user.

By screening out new registered users or users who have not queried products and matching them with similar users, it is convenient to recommend products based on similar users.

According to the login information of the similar user obtained in step S403, the historical information of the similar user is obtained and used as the historical query information of the target user.

Referring back to fig. 2, in step S203, first associated data is obtained through data dimension reduction processing based on the user history query information.

Fig. 5 schematically illustrates a flowchart of obtaining first associated data through data processing in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 5, the method 500 includes steps S501 to S503.

And step S501, performing a labeling operation on the second associated data in the user history query information based on the second feature data, to obtain labeled third associated data.

For example, the second characteristic data includes gender male, age 25 years, profession sales, hobbies running, and basketball. The second associated data is query information data having second characteristic data. And marking the second associated data to obtain tagged third associated data. The tagged third related data may be tagged query information data with tag name of gender and tag value of male, tagged query information data with tag name of age and tag value of 25 years old, tagged query information data with tag name of occupation and tag value of sales, tagged query information data with tag name of interest and tag value of running and basketball, etc., wherein the tag of the tagged query information data is not limited to one tag, and the same data may have a plurality of tags.

Step S502, performing data cleaning on the third associated data to obtain cleaned fourth associated data, where the data cleaning includes removing abnormal data in the third associated data, and the abnormal data includes incomplete data, error data and repeated data.

And carrying out data cleaning such as data noise reduction on the labeled third associated data, and removing abnormal data such as incomplete data, error data, repeated data and the like through data cleaning to obtain fourth associated data.

Step S503, performing data feature fusion on the fourth associated data to obtain first associated data.

And performing word segmentation operation and semantic analysis on the fourth associated data, and storing and de-duplication processing the data with the same semantic to obtain fused first associated data.

And the data dimension reduction processing is performed through tagging, data decontamination and data fusion, so that the computer resource is saved.

Referring back to fig. 2, in step S204, a preference value of the user for the product is obtained through a preference value generating model according to the first association data, wherein the preference value generating model is obtained through pre-training.

Fig. 6 schematically illustrates a flowchart of a training preference value generation model in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 6, the method 600 includes steps S601 to S606.

In step S601, a training data set is acquired, the training data set including first association data for training.

Step S602, based on the first association data for training, implicit characteristics common to users and financial products in the training data set are obtained.

For example, user u and product parameter c have multiple implicit features k in common.

Step S603, obtaining a first weight of the user on the implicit feature, and constructing a first weight matrix of the user on the implicit feature.

For example, p _u，k The relation between the preference of the user u and the f hidden class is measured, namely the weight value of the user u to the hidden characteristic k.

Constructing a first weight matrix of user u for implicit feature k can be expressed as:

u\k	k1	k2	k3
				u1	p _u1，k1	p _u1，k2	p _u1，k3
u2	p _u2，k1	p _u2，k2	p _u2，k3
				u3	p _u3，k1	p _u3，k2	p _u3，k3

for example, the implicit characteristic k1 of the user u1 is small in investment, and the weight is set to be 0.7; the hidden characteristic k2 of the user u1 is low in risk, and the weight is set to be 0.6; the implicit characteristic k3 of the user u1 is high in benefit, and the weight is set to be 0.1. The implicit characteristic k1 of the user u2 is small in investment, and the weight is set to be 0.8; the hidden characteristic k2 of the user u2 is low in risk, and the weight is set to be 0.4; the implicit characteristic k3 of the user u2 is high in benefit, and the weight is set to be 0.6. The implicit characteristic k1 of the user u3 is small in investment, and the weight is set to be 0.5; the hidden characteristic k2 of the user u3 is low in risk, and the weight is set to be 0.6; the implicit characteristic k3 of the user u3 is high in benefit, and the weight is set to be 0.3.

The first weight matrix is expressed as:

u\k	investment is small	Low risk	High profit
				User 1	0.7	0.6	0.1
User 2	0.8	0.4	0.6
				User 3	0.5	0.6	0.3

The above examples merely exemplify the user and the implicit feature, and are not limited thereto, and may be a plurality of users and a plurality of implicit features.

Step S604, obtaining a second weight of the financial product to the implicit characteristic, and constructing a second weight matrix of the financial product to the implicit characteristic.

For example, q _c，k The relation between the product c and the f hidden class is measured, namely the weight value of the product parameter c to the hidden characteristic k.

Constructing a second weight matrix for product c versus implicit feature k can be expressed as:

c\k	k1	k2	k3
				c1	q _c1，k1	q _c1，k2	q _c1，k3
c2	q _c2，k1	q _c2，k2	q _c2，k3
				c3	q _c3，k1	q _c3，k2	q _c3，k3

for example, the implicit characteristic k1 of the product c1 is small investment, and the weight is set to be 0.8; the hidden characteristic k2 of the product c1 is low in risk, and the weight is set to be 0.4; the implicit characteristic k3 of the product c1 is high in benefit, and the weight is set to be 0.2. The implicit characteristic k1 of the product c2 is small investment, and the weight is set to be 0.6; the hidden characteristic k2 of the product c2 is low in risk, and the weight is set to be 0; the implicit characteristic k3 of the product c2 is high in benefit, and the weight is set to be 0.7. The implicit characteristic k1 of the product c3 is small investment, and the weight is set to be 0.5; the hidden characteristic k2 of the product c3 is low in risk, and the weight is set to be 0.7; the implicit characteristic k3 of the product c3 is high in benefit, and the weight is set to be 0.6.

The second weight matrix is expressed as:

c\k	investment is small	Low risk	High profit
				Product 1	0.8	0.4	0.2
Product 2	0.6	0	0.7
				Product 3	0.5	0.7	0.6

The above examples merely exemplify products and implicit features, but are not limited thereto, and may be a plurality of products and a plurality of implicit features.

Step S605, based on the implicit characteristic, fuses the first weight matrix and the second weight matrix to obtain a preference value matrix of the user for the financial product.

And fusing the first weight matrix and the second weight matrix to obtain a preference value matrix of the user on the financial product.

The matrix of preference values of the user for the financial product is expressed as:

wherein, the liquid crystal display device comprises a liquid crystal display device,

according to the fusion method, the first weight matrix and the second weight matrix are fused into a preference value matrix of the user on the financial product, and the preference value matrix is expressed as follows:

the above examples are merely examples of users and products, and are not limited thereto, and may be a plurality of products and a plurality of product features.

Step S606, calculating the preference value of the user to the financial product according to the preference value matrix of the user to the financial product.

Because the obtained preference value matrix of the user for the financial product is discrete evidence when the preference value of the user for the financial product is actually obtained, the calculated amount is huge, and the upper limit F of F is set, so that the calculation is ensured to be successfully completed.

Calculating a preference value of a user for a financial product by the following formula:

wherein p is _u，k The relation between the preference of the user u and the f hidden class is measured, namely the weight value of the user u to the hidden characteristic k, and q _c，k The relation between the product parameter c and the F hidden type is measured, namely the weight value of the product parameter c to the hidden characteristic k, T is a transposition function, and F is the maximum value of F.

And the preference value of the user and the product is determined through the hidden relation model, so that the product is conveniently recommended to the user, and the hidden relation model is constructed through the matrix, thereby being beneficial to improving the accuracy of the preference value.

Fig. 7 schematically illustrates a flowchart of a verification preference value generation model in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 7, the method 700 includes steps S701 to S704.

In step S701, a test data set is acquired, where the test data set includes first association data for a test.

In step S702, a test data set verification success rate threshold is preset.

Step S703, verifying the data in the test data set, and obtaining a verification success rate.

Fig. 8 schematically illustrates a flowchart for verifying data in a test data set in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 8, the method 800 includes steps S801 to S805.

Step S801, obtaining a real preference value of a test user to a test product according to the first association data for testing.

Step S802, inputting the test user into the hidden relation model to obtain a test preference value of the test product.

And step 803, performing loss optimization on the real preference value and the test preference value in a mean square error mode to obtain a loss value.

The smaller the loss function value, the closer the values of the corresponding predicted and actual results are.

In step S804, a loss threshold is preset.

Step S805, if the loss value is within the loss threshold, the verification is successful.

Referring back to fig. 7, in step S704, when the verification success rate is greater than the success rate threshold, the hidden-type relationship model is determined.

The hidden relation model is verified through the success rate threshold value and the test success rate of the test set, the accuracy of the model is improved, and the test data is verified in a mode of loss optimization through mean square error, so that the accuracy of the model is improved.

Referring back to fig. 2, in step S205, a recommendation list including a plurality of recommended financial products is determined according to the preference values.

Step S206, based on the recommendation list, obtaining product parameters and dynamic information corresponding to a plurality of recommended financial products in the recommendation list, where the dynamic information includes: financial product browsing amount, financial product attention amount, financial product transaction amount and financial product evaluation.

Step S207, generating dynamic data of the financial product according to the product parameters and the dynamic information.

For example, dynamic data of a financial product may be based on a financial product browsing volume, a financial product interest volume, a financial product transaction volume, and a financial product evaluation animation clip. For example, the transaction amount of a certain financial product is obviously increased in the last half year, and the appreciation is increased. The reasons are described in the animation stub to make it easier for the customer to learn about the financial product.

Step S208, outputting a plurality of recommended financial products and dynamic data in the recommendation list.

Fig. 9 schematically illustrates a flowchart of a user querying a financial product in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 9, in the method 900, a user enters a product parameter query system, the system determines whether the user is a registered user, if the user is a registered user, the user is allowed to log in directly, if the user is an unregistered user, or the user is registered and logged in again by skipping to a registration page. After the user logs in, the system jumps to the search interface and recommends financial products and dynamic data to the user. The user can inquire according to the currently recommended products, and can search for favorite products by searching, and meanwhile, the computer enters a database to retrieve product data for display.

Fig. 10 schematically illustrates a flowchart of information interaction when a user inquires about a financial product in a financial product recommendation method according to an embodiment of the present disclosure.

As shown in fig. 10, in the method 1000, after a client initiates a query request to a back-end server, the back-end server initiates a query information request to a product parameter information database, the product parameter information database returns data to the back-end server, and the back-end server returns data to the client.

By the financial product recommendation method, the calculation amount of a computer is reduced by data dimension reduction, and the calculation efficiency is improved; moreover, the preference value of the user and the product is determined through the hidden relation model, so that the accuracy of recommending the product is improved; and the customers without professional financial knowledge can easily know the financial products through the display of the dynamic data, so that the financial products suitable for the customers can be selected.

The financial product recommendation device of the disclosed embodiment will be described in detail below with reference to fig. 11 to 15 based on the scenario described in fig. 1. It should be noted that the above application scenario is only shown for the convenience of understanding the spirit and principles of the present disclosure, and the embodiments of the present disclosure are not limited in any way in this respect. Rather, embodiments of the present disclosure may be applied to any scenario where applicable.

Fig. 11 schematically illustrates a block diagram of a financial product recommendation device according to an embodiment of the present disclosure.

As shown in fig. 11, the financial product recommendation device 1100 includes: a first acquisition module 1101, a second acquisition module 1102, a data processing module 1103, a model generation module 1104, a first determination module 1105, a third acquisition module 1106, a data generation module 1107, and an output module 1108.

The first obtaining module 1101 is configured to obtain user login information. In an embodiment, the first obtaining module 1101 may be used to perform the step S201 described above, which is not described herein.

The second obtaining module 1102 is configured to obtain user history query information in response to the user login information. In an embodiment, the second obtaining module 1102 may be configured to perform step S202 described above.

Fig. 12 schematically illustrates a block diagram of a second acquisition module in a financial product recommendation device according to an embodiment of the present disclosure.

As shown in fig. 12, the second acquisition module 1102 includes: a judgment module 1201, a first comparison module 1202 and a fourth acquisition module 1203.

A determining module 1201 is configured to determine whether the user history query information is empty. In an embodiment, the determining module 1201 may be used to perform the step S301 described above, which is not described herein.

The first comparison module 1202 is configured to obtain historical user login information if the historical query information of the user is empty, and compare the user login information with the historical user login information in similarity to obtain login information of a similar user. In an embodiment, the first comparison module 1202 may be configured to perform step S302 described above.

Fig. 13 schematically illustrates a block diagram of a first contrast module in a financial product recommendation device according to an embodiment of the present disclosure.

As shown in fig. 13, the first contrast module 1202 includes: a fifth acquisition module 1301, a sixth acquisition module 1302, and a second comparison module 1303.

A fifth obtaining module 1301, configured to obtain first feature data of the user login information. In an embodiment, the fifth obtaining module 1301 may be configured to perform the step S401 described above, which is not described herein.

A sixth obtaining module 1302, configured to obtain second feature data of the historical user login information. In an embodiment, the sixth obtaining module 1302 may be configured to perform the step S402 described above, which is not described herein.

The second comparing module 1303 is configured to compare the first feature data with the second feature data, and take historical user login information of the second feature data that is the same as the first feature data as login information of a similar user. In an embodiment, the second comparing module 1303 may be used to perform the step S403 described above, which is not described herein.

Referring back to fig. 12, the fourth obtaining module 1203 is configured to obtain, according to the login information of the similar user, historical query information of the similar user as the historical query information of the user. In an embodiment, the fourth obtaining module 1203 may be configured to perform the step S303 described above, which is not described herein.

Referring back to fig. 11, the data processing module 1103 is configured to obtain the first associated data through a data dimension reduction process based on the user history query information.

Fig. 14 schematically illustrates a block diagram of a data processing module in a financial product recommendation device according to an embodiment of the present disclosure.

As shown in fig. 14, the data processing module 1103 includes: a tagging module 1401, a data cleansing module 1402, and a data fusion module 1403.

The labeling module 1401 is configured to perform a labeling operation on the second associated data in the user history query information based on the second feature data, so as to obtain labeled third associated data. In an embodiment, the labeling module 1401 may be used to perform the step S501 described above, which is not described herein.

The data cleansing module 1402 is configured to perform data cleansing on the third associated data to obtain cleansed fourth associated data, where the data cleansing includes removing abnormal data in the third associated data, and the abnormal data includes incomplete data, erroneous data, and repeated data. In an embodiment, the data cleansing module 1402 may be used to perform the step S502 described above, which is not described herein.

The data fusion module 1403 is configured to perform data feature fusion on the fourth associated data to obtain first associated data. In an embodiment, the data fusion module 1403 may be used to perform the step S503 described above, which is not described herein.

Referring back to fig. 11, the model generating module 1104 is configured to obtain a preference value of the user for the product through a preference value generating model according to the first association data, where the preference value generating model is obtained through pre-training.

Fig. 15 schematically illustrates a block diagram of a model generation module of a financial product recommendation device according to an embodiment of the present disclosure.

As shown in fig. 15, the model generation module 1104 includes: a seventh acquisition module 1501, an eighth acquisition module 1502, a first matrix construction module 1503, a second matrix construction module 1504, a matrix fusion module 1505, a calculation module 1506, a ninth acquisition module 1507, a first preset module 1508, a tenth acquisition module 1509, and a second determination module 1510.

A seventh obtaining module 1501 is configured to obtain a training data set, where the training data set includes first association data for training. In an embodiment, the seventh obtaining module 1501 may be configured to perform the step S601 described above, which is not described herein.

An eighth obtaining module 1502 is configured to obtain implicit characteristics common to the user and the financial product in the training dataset based on the training first association data. In an embodiment, the eighth obtaining module 1502 may be configured to perform the step S602 described above, which is not described herein.

The first matrix construction module 1503 is configured to obtain a first weight of the implicit feature by the user, and construct a first weight matrix of the implicit feature by the user. In an embodiment, the first matrix construction module 1503 may be used to perform the step S603 described above, which is not described herein.

A second matrix construction module 1504 is configured to acquire a second weight of the financial product on the implicit feature, and construct a second weight matrix of the financial product on the implicit feature. In an embodiment, the second matrix construction module 1504 may be used to perform the step S604 described above, which is not described herein.

The matrix fusion module 1505 is configured to fuse the first weight matrix and the second weight matrix based on the implicit feature, so as to obtain a preference value matrix of the user for the financial product. In an embodiment, the matrix fusion module 1505 may be used to perform the step S605 described above, which is not described herein.

The calculating module 1506 is configured to calculate a preference value of the user for the financial product according to the preference value matrix of the user for the financial product. In an embodiment, the calculating module 1506 may be configured to perform the step S606 described above, which is not described herein.

A ninth obtaining module 1507 is configured to obtain a test data set, where the test data set includes first association data for a test. In an embodiment, the ninth obtaining module 1507 may be used to perform the step S701 described above, which is not described herein.

A preset module 1508 for presetting a test data set verification success rate threshold. In an embodiment, the presetting module 1508 may be used to perform the step S702 described above, which is not described herein.

A tenth obtaining module 1509 is configured to verify the data in the test data set, and obtain a verification success rate. In an embodiment, the tenth acquiring module 1509 may be used to perform the step S703 described above, which is not described herein.

A second determining module 1510 is configured to determine the hidden relation model when the verification success rate is greater than the success rate threshold. In an embodiment, the second determining module 1510 may be used to perform the step S704 described above, which is not described herein.

Referring back to fig. 11, the first determining module 1105 is configured to determine a recommendation list including a plurality of recommended financial products according to the preference value. In an embodiment, the first determining module 1105 may be configured to perform the step S205 described above, which is not described herein.

A third obtaining module 1106 is configured to obtain, based on the recommendation list, product parameters and dynamic information corresponding to a plurality of recommended financial products in the recommendation list. In an embodiment, the third obtaining module 1106 may be configured to perform the step S206 described above, which is not described herein.

A data generating module 1107, configured to generate dynamic data of the financial product according to the product parameter and the dynamic information. In an embodiment, the data generating module 1107 may be configured to perform the step S207 described above, which is not described herein.

An output module 1108 is configured to output the plurality of recommended financial products and the dynamic data in the recommendation list. In an embodiment, the output module 1108 may be used to perform the step S208 described above, which is not described herein.

According to an embodiment of the present disclosure, any of the first acquisition module 1101, the second acquisition module 1102, the data processing module 1103, the model generation module 1104, the first determination module 1105, the third acquisition module 1106, the data generation module 1107, and the output module 1108 may be combined in one module to be implemented, or any of them may be split into a plurality of modules. Alternatively, at least some of the functionality of one or more of the modules may be combined with at least some of the functionality of other modules and implemented in one module. According to embodiments of the present disclosure, at least one of the first acquisition module 1101, the second acquisition module 1102, the data processing module 1103, the model generation module 1104, the first determination module 1105, the third acquisition module 1106, the data generation module 1107, and the output module 1108 may be implemented at least in part 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 a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in hardware or firmware in any other reasonable manner of integrating or packaging circuitry, or in any one of or a suitable combination of three of software, hardware, and firmware implementations. Alternatively, at least one of the first acquisition module 1101, the second acquisition module 1102, the data processing module 1103, the model generation module 1104, the first determination module 1105, the third acquisition module 1106, the data generation module 1107 and the output module 1108 may be at least partially implemented as computer program modules which, when executed, may perform the respective functions.

As shown in fig. 16, an electronic device 1200 according to an embodiment of the present disclosure includes a processor 1601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 1602 or a program loaded from a storage section 1608 into a Random Access Memory (RAM) 1603. The processor 1601 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)), or the like. The processor 1601 may also include on-board memory for caching purposes. The processor 1601 may include a single processing unit or multiple processing units for performing different actions in accordance with the method flows of the disclosed embodiments.

In the RAM1603, various programs and data necessary for the operation of the electronic apparatus 1200 are stored. The processor 1601, ROM1602, and RAM1603 are connected to each other by a bus 1604. The processor 1601 performs various operations of the method flow according to the embodiments of the present disclosure by executing programs in the ROM1602 and/or RAM 1603. Note that the program can also be stored in one or more memories other than the ROM1602 and the RAM 1603. The processor 1601 may also perform various operations of the method flow according to an embodiment of the present disclosure by executing programs stored in the one or more memories.

According to an embodiment of the disclosure, the electronic device 1200 may also include an input/output (I/O) interface 1605, the input/output (I/O) interface 1605 also being connected to the bus 1604. The electronic device 1200 may also include one or more of the following components connected to the I/O interface 1605: an input portion 1606 including a keyboard, a mouse, and the like; an output portion 1607 including a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker, and the like; a storage section 1608 including a hard disk or the like; and a communication section 1609 including a network interface card such as a LAN card, a modem, or the like. The communication section 1609 performs communication processing via a network such as the internet. The drive 1610 is also connected to the I/O interface 1605 as needed. A removable medium 1611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is installed as needed on the drive 1610 so that a computer program read out therefrom is installed into the storage section 1608 as needed.

The present disclosure also provides a computer-readable storage medium that may be embodied in the apparatus/device/system described in the above embodiments; or may exist alone without being assembled into the apparatus/device/system. The computer-readable storage medium carries one or more programs which, when executed, implement methods in accordance with embodiments of the present disclosure.

According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example, but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, the computer-readable storage medium may include ROM1602 and/or RAM1603 described above and/or one or more memories other than ROM1602 and RAM 1603.

Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowcharts. The program code, when executed in a computer system, causes the computer system to implement the financial product recommendation method provided by embodiments of the present disclosure.

The above-described functions defined in the system/apparatus of the embodiments of the present disclosure are performed when the computer program is executed by the processor 1601. The systems, apparatus, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the disclosure.

In one embodiment, the computer program may be based on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program can also be transmitted, distributed over a network medium in the form of signals, downloaded and installed via the communication portion 1609, and/or from the removable medium 1611. The computer program may include program code that may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

In such embodiments, the computer program may be downloaded and installed from a network via the communication portion 1609, and/or installed from the removable media 1611. The above-described functions defined in the system of the embodiments of the present disclosure are performed when the computer program is executed by the processor 1601. The systems, devices, apparatus, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the disclosure.

According to embodiments of the present disclosure, program code for performing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, such computer programs may be implemented in high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages. Programming languages include, but are not limited to, such as Java, c++, python, "C" or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Those skilled in the art will appreciate that the features recited in the various embodiments of the disclosure and/or in the claims may be provided in a variety of combinations and/or combinations, even if such combinations or combinations are not explicitly recited in the disclosure. In particular, the features recited in the various embodiments of the present disclosure and/or the claims may be variously combined and/or combined without departing from the spirit and teachings of the present disclosure. All such combinations and/or combinations fall within the scope of the present disclosure.

The embodiments of the present disclosure are described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the embodiments cannot be used advantageously in combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be made by those skilled in the art without departing from the scope of the disclosure, and such alternatives and modifications are intended to fall within the scope of the disclosure.

Claims

1. A method of recommending a financial product, the method comprising:

acquiring user login information;

2. The method of claim 1, wherein obtaining the user history query information in response to user login information comprises:

judging whether the historical query information of the user is empty or not;

3. The method of claim 2, wherein comparing the user login information with historical user login information for similarity, the obtaining login information for a similar user comprises:

acquiring first characteristic data of the user login information;

4. The method of claim 3, wherein obtaining first association data by data processing based on the user history query information comprises:

5. The method of claim 1, wherein the preference value generation model is a hidden class relationship model, and pre-training the preference value generation model comprises:

6. The method of claim 5, wherein pre-training the preference value generation model further comprises:

Presetting a test data set verification success rate threshold;

wherein verifying the data in the test dataset comprises:

presetting a loss threshold; and

if the loss value is within the loss threshold, the verification is successful.

7. A financial product recommendation device, the device comprising:

the first acquisition module is used for acquiring user login information;

8. An electronic device, comprising:

one or more processors;

storage means for storing one or more programs,

wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-6.

9. A computer readable storage medium having stored thereon executable instructions which when executed by a processor cause the processor to perform the method of any of claims 1 to 6.

10. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 6.