CN114202380A

CN114202380A - Recommendation method, device and equipment for financial products

Info

Publication number: CN114202380A
Application number: CN202111488110.3A
Authority: CN
Inventors: 曹清鑫
Original assignee: China Construction Bank Corp
Current assignee: China Construction Bank Corp
Priority date: 2021-12-07
Filing date: 2021-12-07
Publication date: 2022-03-18

Abstract

The application provides a recommendation method, device and equipment of financial products, and relates to the technical field of recommendation systems, wherein the method comprises the steps of obtaining a to-be-processed financial product list, wherein the to-be-processed financial product list comprises at least one to-be-processed financial product; determining product label information of the financial products to be processed in the financial product list to be processed, and acquiring key information of information corresponding to the user; inputting product label information and key information of the financial products to be processed in the financial product list to be processed into a preset recommendation model to obtain a financial product list to be recommended; and recommending the financial products to be recommended to the user. Determining financial products which are interested by the user by combining the product label information of the financial products and the information browsed by the user, and recommending the financial products for the user by combining various characteristics; financial products can be accurately recommended to the user.

Description

Recommendation method, device and equipment for financial products

Technical Field

The application relates to the technical field of recommendation systems, in particular to a recommendation method, device and equipment for financial products.

Background

With the development of information technology, financial products can be analyzed based on the information technology. Financial products may be recommended to a user based on information technology.

In the prior art, financial products can be recommended to a user based on the situation that the user browses the financial products historically.

However, in the prior art, the financial products are recommended to the user only based on the situation that the user browses the financial products historically, which is not accurate; financial products matched with the user cannot be accurately recommended to the user, and if the user does not browse the financial products, the financial products cannot be recommended to the user.

Disclosure of Invention

The application provides a method, a device and equipment for recommending financial products, which are used for solving the problem that financial products matched with a user cannot be accurately recommended to the user.

In a first aspect, the present application provides a method for recommending financial products, the method comprising:

acquiring a financial product list to be processed, wherein the financial product list to be processed comprises at least one financial product to be processed;

determining product label information of the financial products to be processed in the financial product list to be processed, and acquiring key information of information corresponding to the user;

inputting product label information of the financial products to be processed in the financial product list to be processed and key information of the information into a preset recommendation model to obtain a financial product list to be recommended, wherein the financial product list to be recommended comprises at least one financial product to be recommended;

and recommending the financial product to be recommended to the user.

In one possible implementation, the product tag information includes one or more of the following: the product category, the risk level, the product data information, the popularity in the first time period and the popularity in the second time period; wherein the time length of the first time period is greater than the time length of the second time period.

In one possible implementation manner, if the product tag information includes a popularity in a first time period and a popularity in a second time period, determining the product tag information of the financial product to be processed in the list of financial products to be processed includes:

acquiring each piece of first product data information of the financial product to be processed in the financial product list to be processed in the first time period and each piece of second product data information of the financial product to be processed in the second time period, determining the weight corresponding to each piece of first product data information of the financial product to be processed in the first time period, and determining the weight corresponding to each piece of second product data information of the financial product to be processed in the second time period;

determining the popularity of the financial product to be processed in the first time period according to the first product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of first product data information;

and determining the popularity of the financial product to be processed in the second time period according to the second product data information of the financial product to be processed in the first time period and the weight corresponding to each second product data information.

In one possible implementation manner, the obtaining key information of the information corresponding to the user includes:

and acquiring the information browsed by the user, and extracting key information of the information by adopting a natural language processing mode.

In a possible implementation manner, before inputting the product tag information of the financial product to be processed in the financial product list to be processed and the key information of the information into a preset recommendation model to obtain the financial product list to be recommended, the method further includes:

obtaining an interest tag of the user, and determining a financial product to be processed matched with the interest tag in the financial product list to be processed;

and the financial products to be processed matched with the interest tags are the financial products input into a preset recommendation model.

In a possible implementation manner, the method further includes:

acquiring a financial product list to be trained, wherein the financial product list to be trained comprises at least one financial product to be trained;

determining product label information of financial products to be processed in the financial product list to be trained, and acquiring information browsed by a user and user interest label information;

performing feature extraction on the product label information of the financial product to be processed and information browsed by a user to obtain a feature vector; training an initial model according to the feature vector to obtain an initial recommendation model;

and performing incremental iterative training on the initial recommendation model according to the product label information of the financial product to be processed and the user interest label information to obtain a recommendation model for obtaining the financial product to be recommended.

In one possible implementation manner, if the product tag information includes a popularity in a first time period and a popularity in a second time period, where a time length of the first time period is greater than a time length of the second time period, determining the product tag information of the financial product to be processed in the financial product list to be trained includes:

acquiring each piece of first product data information of the financial product to be trained in the financial product list to be trained in the first time period and each piece of second product data information of the financial product to be trained in the second time period, determining the weight corresponding to each piece of first product data information of the financial product to be trained in the first time period, and determining the weight corresponding to each piece of second product data information of the financial product to be trained in the second time period;

determining the popularity of the financial product to be trained in the first time period according to the first product data information of the financial product to be trained in the first time period and the weight corresponding to each piece of first product data information;

and determining the popularity of the financial product to be trained in the second time period according to the second product data information of the financial product to be trained in the first time period and the weight corresponding to each second product data information.

In a possible implementation manner, the user interest tag information includes a user interest level in a first time period and a user interest level in a second time period, wherein a time length of the first time period is greater than a time length of the second time period; acquiring user interest tag information, comprising:

acquiring first behavior data of a user on a financial product in a first time period and second behavior data of the user on the financial product in a second time period;

determining the interest degree of the user in a first time period according to each first behavior data and the weight corresponding to each first behavior data; and determining the interest degree of the user in the second time period according to each second behavior data and the weight corresponding to each second behavior data.

In a second aspect, the present application provides an apparatus for recommending financial products, the apparatus comprising:

the device comprises a first acquisition unit, a second acquisition unit and a processing unit, wherein the first acquisition unit is used for acquiring a financial product list to be processed, and the financial product list to be processed comprises at least one financial product to be processed;

a first determining unit for determining product tag information of the financial product to be processed in the list of financial products to be processed;

the second acquisition unit is used for acquiring key information of the information corresponding to the user;

the second determining unit is used for inputting the product label information of the financial product to be processed in the financial product list to be processed and the key information of the information into a preset recommendation model to obtain a financial product list to be recommended, wherein the financial product list to be recommended comprises at least one financial product to be recommended;

and the recommending unit is used for recommending the financial product to be recommended to the user.

In a possible implementation manner, if the product tag information includes a popularity in the first time period and a popularity in the second time period, the first determining unit includes:

a first obtaining module, configured to obtain first product data information of a financial product to be processed in the financial product list to be processed in the first time period and second product data information of the financial product to be processed in the second time period, determine a weight corresponding to each first product data information of the financial product to be processed in the first time period, and determine a weight corresponding to each second product data information of the financial product to be processed in the second time period;

the first determining module is used for determining the popularity of the financial product to be processed in the first time period according to the first product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of first product data information;

and the second determining module is used for determining the popularity of the financial product to be processed in the second time period according to the second product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of second product data information.

In a possible implementation manner, the second obtaining unit is specifically configured to:

In a possible implementation manner, the apparatus further includes:

a third obtaining unit, configured to obtain an interest tag of the user before the second determining unit inputs product tag information of a financial product to be processed in the financial product list to be processed and key information of the information into a preset recommendation model to obtain a financial product list to be recommended, and determine a financial product to be processed in the financial product list to be processed, where the financial product to be processed is matched with the interest tag;

In a possible implementation manner, the apparatus further includes:

the training unit is used for acquiring a financial product list to be trained, and the financial product list to be trained comprises at least one financial product to be trained;

determining product label information of financial products to be processed in the financial product list to be trained;

acquiring information browsed by a user;

acquiring user interest tag information;

In a possible implementation manner, if the product tag information includes a popularity in a first time period and a popularity in a second time period, the training unit, when determining the product tag information of the financial product to be processed in the financial product list to be trained, is specifically configured to:

In a possible implementation manner, the user interest tag information includes a user interest level in a first time period and a user interest level in a second time period, wherein a time length of the first time period is greater than a time length of the second time period; when obtaining the user interest tag information, the training unit is specifically configured to:

In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

a memory; a memory for storing the processor-executable instructions;

wherein the processor is configured to perform the method of the first aspect.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, the computer-executable instructions are used to implement the method of the first aspect.

In a fifth aspect, an embodiment of the present application provides a computer program product, where the computer program product includes: a computer program, stored in a readable storage medium, from which at least one processor of an electronic device can read the computer program, execution of the computer program by the at least one processor causing the electronic device to perform the method of the first aspect.

The method, the device and the equipment for recommending the financial products acquire a to-be-processed financial product list, wherein the to-be-processed financial product list comprises at least one to-be-processed financial product; determining product label information of the financial products to be processed in the financial product list to be processed, and acquiring key information of information corresponding to the user; inputting product label information and key information of the information of financial products to be processed in a financial product list to be processed into a preset recommendation model to obtain a financial product list to be recommended, wherein the financial product list to be recommended comprises at least one financial product to be recommended; and recommending the financial products to be recommended to the user. Combining product label information of the financial products and information browsed by the user (for example, the information browsed by the user and information similar to the information browsed by the user), determining the financial products which the user is interested in, and combining various characteristics to recommend the financial products for the user; financial products can be accurately recommended to the user.

Drawings

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

Fig. 1 is a first scenario diagram provided in an embodiment of the present application;

fig. 2 is a schematic view of a scenario two provided in the embodiment of the present application;

FIG. 3 is a flowchart illustrating a method for recommending financial products according to an embodiment of the present disclosure;

FIG. 4 is a schematic flow chart illustrating another method for recommending financial products according to an embodiment of the present disclosure;

FIG. 5 is an interface diagram of another method for recommending financial products according to an embodiment of the present application;

FIG. 6 is a schematic structural diagram of an apparatus for recommending financial products according to an embodiment of the present application;

FIG. 7 is a schematic structural diagram of another apparatus for recommending financial products according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure;

FIG. 9 is a block diagram illustrating an electronic device in accordance with an example embodiment.

With the above figures, there are shown specific embodiments of the present application, which will be described in more detail below. These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like 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 application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

In one example, financial products may be recommended to a user based on a user's historical viewing of financial products. For example, if the user browses financial products of the insurance category, the financial products under the insurance category are recommended to the user; and if the user browses the financial products of the financing category, recommending the financial products under the financing category to the user.

However, in the above manner, the recommendation of the financial product to the user is not accurate only based on the situation that the user browses the financial product historically; financial products matched with the user cannot be accurately recommended to the user, and if the user does not browse the financial products, the financial products cannot be recommended to the user.

The application provides a recommendation method, device and equipment for financial products, and aims to solve the above technical problems in the prior art.

The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.

Fig. 1 is a first scenario diagram provided in the embodiment of the present application, as shown in fig. 1, when a user uses a terminal device, a recommended financial product may be displayed through an interface of an application program of the terminal device. For example, financial product 1, financial product 2, and financial product 3 are recommended to the user, and the detailed information of financial product 1, the detailed information of financial product 2, and the detailed information of financial product 3 are displayed.

FIG. 2 is a schematic view of a second scenario provided by an embodiment of the present application, as shown in FIG. 2, when a user enters a business hall, a person in the business hall may recommend a financial product for the user based on the user; the recommended financial product is displayed on a tablet computer used by a person in the business hall. For example, financial product 1, financial product 2, and financial product 3 are recommended to the user, and the detailed information of financial product 1, the detailed information of financial product 2, and the detailed information of financial product 3 are displayed.

Fig. 3 is a schematic flowchart of a method for recommending a financial product according to an embodiment of the present application, where as shown in fig. 3, the method includes:

301. the method comprises the steps of obtaining a to-be-processed financial product list, wherein the to-be-processed financial product list comprises at least one to-be-processed financial product.

Illustratively, a list of financial products to be processed may first be obtained. The method comprises the steps that a financial product list to be processed can be obtained from a local database, wherein the database comprises a plurality of financial products; alternatively, the list of financial products to be processed is obtained from another device.

The financial product list to be processed comprises a plurality of financial products to be processed; in the present embodiment, a financial product recommended to the user is determined from these financial products to be processed.

302. Determining the product label information of the financial product to be processed in the financial product list to be processed, and acquiring the key information of the corresponding information of the user.

Illustratively, since the financial product has product information, e.g., risk level, product type, term, rate of return, purchase amount, etc.; the product information may be determined as product tag information of the financial product. And further obtaining product label information of each financial product to be processed in the financial product list to be processed, namely the product label information comprises the following label information: product category, risk level, product data information, etc.; the product data information includes, among other things, terms, profitability, purchase amount, etc.

The method and the system for recommending the financial products combine information browsed by the user to recommend the financial products. Information browsed by the user and information similar to the information browsed by the user can be obtained. Wherein each information has a title, a subject content, and a summary content. Keywords for each piece of information can be extracted based on natural language processing.

The title, subject content, content synopsis, and keywords of the information constitute the key information of the information.

Different weights may also be assigned to different key information, for example, a weight may be assigned to a title, a weight may be assigned to a subject content, a weight may be assigned to a summary of content, and a weight may be assigned to a keyword. For each piece of information, the information has a plurality of pieces of key information, different pieces of key information have different weights, and the information tag value of the information can be obtained by performing weighted summation on each piece of key information and the weight corresponding to each piece of key information; obtaining the label of the information according to the corresponding relationship between the different information label values and the labels of the information. For example, the label of the information is endowment insurance, endowment, insurance, investment, and financing.

303. And inputting the product label information and the key information of the financial products to be processed in the financial product list to be processed into a preset recommendation model to obtain the financial product list to be recommended, wherein the financial product list to be recommended comprises at least one financial product to be recommended.

Illustratively, the pre-training results in a recommendation model that is trained in conjunction with the consultations viewed by the other users, the financial products of interest to the other users, and the interests of the other users.

And inputting the product label information and the key information of the financial products to be processed in the financial product list to be processed into a preset recommendation model, and outputting to obtain the financial product list to be recommended.

304. And recommending the financial products to be recommended to the user.

Illustratively, displaying the obtained financial products to be recommended to the user; or displaying the obtained financial products to be recommended on a webpage; or sending the obtained financial products to be recommended to the terminal of the user in a short message mode.

In this embodiment, a to-be-processed financial product list is obtained, where the to-be-processed financial product list includes at least one to-be-processed financial product; determining product label information of the financial products to be processed in the financial product list to be processed, and acquiring key information of information corresponding to the user; inputting product label information and key information of the information of financial products to be processed in a financial product list to be processed into a preset recommendation model to obtain a financial product list to be recommended, wherein the financial product list to be recommended comprises at least one financial product to be recommended; and recommending the financial products to be recommended to the user. Combining product label information of the financial products and information browsed by the user (for example, the information browsed by the user and information similar to the information browsed by the user), determining the financial products which the user is interested in, and combining various characteristics to recommend the financial products for the user; financial products can be accurately recommended to the user.

Fig. 4 is a schematic flowchart of another method for recommending financial products according to an embodiment of the present application, where as shown in fig. 4, the method includes:

401. the method comprises the steps of obtaining a financial product list to be trained, wherein the financial product list to be trained comprises at least one financial product to be trained.

Illustratively, a list of financial products to be trained needs to be obtained first. The method comprises the steps that a financial product list to be trained can be obtained from a local database, wherein the database comprises a plurality of financial products; alternatively, the list of financial products to be trained is obtained from other devices.

The financial product list to be trained comprises a plurality of financial products to be trained; in this embodiment, the model is trained according to the financial products to be trained.

402. Determining product label information of the financial products to be processed in the financial product list to be trained.

In one example, if the product label information includes a popularity of the first time period and a popularity of the second time period, wherein the time length of the first time period is greater than the time length of the second time period, step 402 includes the following processes:

the method comprises the steps of obtaining each piece of first product data information of a financial product to be trained in a financial product list to be trained in a first time period and each piece of second product data information of the financial product to be trained in a second time period, determining the weight corresponding to each piece of first product data information of the financial product to be trained in the first time period, and determining the weight corresponding to each piece of second product data information of the financial product to be trained in the second time period.

And determining the popularity of the financial product to be trained in the first time period according to the first product data information of the financial product to be trained in the first time period and the weight corresponding to each piece of first product data information.

Illustratively, since the financial product has product information, e.g., risk level, product type, term, rate of return, purchase amount, etc.; these data can be captured by spark streaming technology or Flink technology. The product information may be determined as product tag information of the financial product. And further obtaining product label information of each financial product to be processed in the financial product list to be processed, namely the product label information comprises the following label information: product category, risk level, product data information, etc.; the product data information includes, among other things, terms, profitability, purchase amount, etc.

And determining the popularity of the financial product in a first time period and the popularity of the financial product in a second time period, wherein the time length of the first time period is longer than that of the second time period. The popularity of the financial product in the first time period is known to be the popularity of the financial product in a long time period; and the popularity of the financial product in the second time period is the popularity of the financial product in a short time.

First product data information for each financial product to be trained over a first time period may be extracted, wherein the first product data information includes the following data information: the financial product is selected from the group consisting of a transaction amount of the financial product in a first time period, a product transaction number of the financial product in the first time period, a viewed number of the financial product in the first time period, a collected number of the financial product in the first time period, a clicked number of the financial product in the first time period, a purchased number of the financial product in the first time period, a number of times the financial product is added to a shopping cart in the first time period, and a focused number of the financial product in the first time period. Different weights may be set for different first product data information.

For example, different weights are set for deal amount, product transaction, browsing, favorites, clicks, purchases, being added to a shopping cart, being attended to. The corresponding weight can be set according to the amount interval to which the transaction amount belongs; setting corresponding weight for the frequency interval to which the product transaction frequency belongs; setting corresponding weight for the frequency interval to which the browsing frequency belongs; setting corresponding weight for the frequency interval to which the collection frequency belongs; setting corresponding weight for the frequency interval to which the click frequency belongs; setting corresponding weight for the frequency interval to which the purchase frequency belongs; setting corresponding weight for the frequency interval to which the frequency of the added shopping cart belongs; and setting corresponding weight for the frequency interval to which the concerned frequency belongs.

And determining the popularity of the financial product to be trained in the first time period according to the first product data information of the financial product to be trained in the first time period and the weight corresponding to each piece of first product data information. For example, the weighting corresponding to each product data information of the financial product to be trained is summed up in a weighted manner, so as to obtain the popularity of the financial product to be trained in the first time period.

For example, the popularity of the first time period for the financial product to be trained is

Wherein i and n are positive integers greater than or equal to 1; n is the type number of the first product data information of the financial product to be trained; x is the number of_iIs the weight of the ith first product data information of the financial product to be trained.

Wherein the first period of time may be 1 day, or 1 month, or 2 months.

Second product data information of each financial product to be trained in a second time period can be extracted, wherein the second product data information comprises the following data information: the transaction amount of the financial product in the second time period, the product transaction times of the financial product in the second time period, the browsed times of the financial product in the second time period, the collected times of the financial product in the second time period, the clicked times of the financial product in the second time period, the purchased times of the financial product in the second time period, the added times of the financial product in the shopping cart in the second time period, and the concerned times of the financial product in the second time period. Different weights may be set for different second product data information.

And determining the popularity of the financial product to be trained in the second time period according to the second product data information of the financial product to be trained in the second time period and the weight corresponding to each second product data information. For example, the weighting corresponding to each product data information of the financial product to be trained is summed up in a weighted manner to obtain the popularity of the financial product to be trained in the second time period.

For example, the popularity of the second time period for taking the financial product to be trained is

Wherein j and m are positive integers greater than or equal to 1; m is the type number of the second product data information of the financial product to be trained; x is the number of_jThe weight of the jth second product data information of the financial product to be trained.

Wherein the second period of time may be 10 days, or half an hour, or 1 hour. The calculation of the popularity in the second time period may be performed by counting the time periods of near N minutes, 3N minutes, M hours, 3M hours, 6M hours, etc. according to the time window of execution. For example, N ═ 10 and M ═ 1.

403. And acquiring information browsed by the user and user interest tag information.

In one example, the user interest tag information comprises the interest degree of the user in a first time period and the interest degree of the user in a second time period, wherein the time length of the first time period is greater than that of the second time period; when obtaining the user interest tag information, the following process may be performed:

acquiring first behavior data of a user on a financial product in a first time period and second behavior data of the user on the financial product in a second time period; determining the interest degree of the user in a first time period according to each first behavior data and the weight corresponding to each first behavior data; and determining the interest degree of the user in the second time period according to each second behavior data and the weight corresponding to each second behavior data.

For example, the present application combines information browsed by a user to recommend a financial product. Information browsed by other users and information similar to the information browsed by other users can be obtained. Wherein each information has a title, a subject content, and a summary content. Keywords for each piece of information can be extracted based on natural language processing.

It is also necessary to acquire user interest tag information of other users. In one example, the behavior data of other users in a long time can be analyzed, and the behavior data of other users in a short time can be analyzed. The behavior data includes: browsing financial products, collecting financial products, clicking on financial products, purchasing financial products, adding financial products to a shopping cart, paying attention to financial products, and so forth.

Acquiring first behavior data of other users on the financial product in a first time period through spark streaming technology or Flink technology, and giving different weights to different first behavior data; and summing the weights of the first behavior data to obtain the interest degree of other users in the first time period. The first period of time is a long period of time. And then obtaining the interest degree of other users in the first time period for each financial product.

For example, the interest level of other users in the first time period is

Wherein F and F are positive integers more than or equal to 1; f is the type number of the first behavior data of the financial products of other users in a first time period; x is the number of_fIs the weight of the f-th first behavior data of the financial product to be trained. Wherein the weights of the different first behavioural data may take different values. Or, the weight is 0 or 1; if the weight of the fth first behavior data is 0, representing that other users do not have the fth first behavior data in the first time period; if the weight of the f-th first behavior data is 1, the f-th first behavior data is represented in the first time period by other users.

Acquiring second behavior data of other users on the financial product in a second time period through spark streaming technology or Flink technology, and giving different weights to different second behavior data; and summing the weights of the second behavior data to obtain the interest degree of other users in the second time period. The first time period is a short time period. And then obtaining the interest degree of other users in the second time period aiming at each financial product.

For example, the interest level of other users in the first time period is

Wherein G and G are positive integers greater than or equal to 1; g is the type quantity of second behavior data of other users on the financial products in a second time period; x is the number of_gIs the weight of the g second behavior data of the financial product to be trained. Wherein the weights of the different second behavior data may take different values. Or, the weight is 0 or 1; if the weight of the g second behavior data is 0, representing that other users do not have the g second behavior data in a second time period; if the weight of the g second behavior data is 1, the g second behavior data of other users is represented in the second time period.

404. Performing feature extraction on product label information of the financial product to be processed and information browsed by a user to obtain a feature vector; and training the initial model according to the feature vector to obtain an initial recommendation model.

Illustratively, a Gradient Boosting Decision Tree (GBDT) is used to extract features of product label information of the financial product to be processed and information browsed by the user, so as to obtain a feature vector.

And training the initial model by the characteristic vector to obtain an initial recommendation model. The initial model may be a Logistic Regression (LR) or a deep learning model or other artificial intelligence model.

In this step, the information browsed by the user includes information browsed by the user in a long time period, information browsed by the user in a short time period, and information similar to the information browsed by the user.

405. And performing incremental iterative training on the initial recommendation model according to the product label information and the user interest label information of the financial product to be processed to obtain the recommendation model for obtaining the financial product to be recommended.

For example, in this embodiment, features may also be extracted from the product tag information and the user interest tag information of the financial product to be processed, the extracted features are subjected to feature splicing, and the features obtained after the feature splicing are input into the initial recommendation model obtained in step 404 to perform incremental iterative training, so as to obtain a recommendation model for obtaining the financial product to be recommended.

In this implementation, the user interest tag information includes the user's interest level for each financial product over a long period of time and the user's interest level for each financial product over a short period of time.

406. The method comprises the steps of obtaining a to-be-processed financial product list, wherein the to-be-processed financial product list comprises at least one to-be-processed financial product.

For example, see step 301, which is not described in detail.

407. Product tag information of the financial product to be processed in the list of financial products to be processed is determined.

In one example, the product label information includes one or more of: the product category, the risk level, the product data information, the popularity in the first time period and the popularity in the second time period; wherein the time length of the first time period is greater than the time length of the second time period.

In one example, if the product label information includes a popularity in the first time period and a popularity in the second time period, step 407 includes the following process:

the method comprises the steps of obtaining each piece of first product data information of a financial product to be processed in a financial product list to be processed in a first time period and each piece of second product data information of the financial product to be processed in a second time period, determining the weight corresponding to each piece of first product data information of the financial product to be processed in the first time period, and determining the weight corresponding to each piece of second product data information of the financial product to be processed in the second time period.

And determining the popularity of the financial product to be processed in the first time period according to the first product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of first product data information.

First product data information for each financial product to be processed may be extracted over a first time period, wherein the first product data information includes the following data information: the financial product is selected from the group consisting of a transaction amount of the financial product in a first time period, a product transaction number of the financial product in the first time period, a viewed number of the financial product in the first time period, a collected number of the financial product in the first time period, a clicked number of the financial product in the first time period, a purchased number of the financial product in the first time period, a number of times the financial product is added to a shopping cart in the first time period, and a focused number of the financial product in the first time period. Different weights may be set for different first product data information.

And determining the popularity of the financial product to be processed in the first time period according to the first product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of the first product data information. For example, the weighting corresponding to each product data information of the financial product to be processed is summed up in a weighted manner, so as to obtain the popularity of the financial product to be processed in the first time period.

For example, the popularity of the first time period of the financial product to be processed is

Wherein Q and Q are positive integers more than or equal to 1; q is the type number of the first product data information of the financial product to be processed; x is the number of_qThe weight of the qth first product data information for the financial product to be processed.

Wherein the first period of time may be 1 day, or 1 month, or 2 months.

Second product data information for each financial product to be processed may be extracted over a second time period, wherein the second product data information includes the following data information: the transaction amount of the financial product in the second time period, the product transaction times of the financial product in the second time period, the browsed times of the financial product in the second time period, the collected times of the financial product in the second time period, the clicked times of the financial product in the second time period, the purchased times of the financial product in the second time period, the added times of the financial product in the shopping cart in the second time period, and the concerned times of the financial product in the second time period. Different weights may be set for different second product data information.

And determining the popularity of the financial product to be processed in the second time period according to the second product data information of the financial product to be processed in the second time period and the weight corresponding to each second product data information. For example, the weighting corresponding to each product data information of the financial product to be processed is summed up in a weighted manner to obtain the popularity of the financial product to be processed in the second time period.

For example, the popularity of the second time period for taking the financial product to be processed is

Wherein H and H are positive integers more than or equal to 1;h is the type and quantity of second product data information of the financial product to be processed; x is the number of_hThe weight of the h-th second product data information of the financial product to be processed.

408. The information browsed by the user is obtained, and the key information of the information is extracted by adopting a natural language processing mode.

For example, refer to step 302, which is not described in detail.

409. Obtaining an interest tag of a user, and determining a financial product to be processed matched with the interest tag in a financial product list to be processed; wherein, the financial product to be processed matched with the interested label is the financial product input into the preset recommendation model.

For example, the financial products to be processed in the financial product list to be processed may be processed by feature intersection based on the interest tag of the user. At this time, the interest tag based on the user may be determined, and the financial product to be processed matched with the interest tag in the list of financial products to be processed is extracted according to the determination of the financial product to be processed matched with the interest tag.

410. And inputting the product label information and the key information of the financial products to be processed in the financial product list to be processed into a preset recommendation model to obtain the financial product list to be recommended, wherein the financial product list to be recommended comprises at least one financial product to be recommended.

Illustratively, the financial products to be processed matched with the interest tags and the key information of the information are input into a preset recommendation model to obtain a list of the financial products to be recommended.

411. And recommending the financial products to be recommended to the user.

Referring to step 304, fig. 5 is an interface diagram of another method for recommending financial products according to an embodiment of the present disclosure, and fig. 5 shows that financial products to be recommended may be displayed on a terminal device; such as financial product 1 to be recommended, financial product 2 to be recommended, financial product 3 to be recommended.

In this embodiment, on the basis of the above embodiment, a recommendation model is obtained by combining product tag information of a financial product to be processed, information browsed by a user, and user interest tag information; and finishing the training process by combining the information browsed by the user to obtain a model accurately recommended for the user. And, the product label information includes a popularity over a long period of time and a popularity over a long period of time; the user interest tag information comprises the interest degree of the user in a long time period for each financial product and the interest degree of the user in a short time period for each financial product; and then the product labels and the interest degree of the user in the financial products are refined in the time dimension, so that a model for preparing the user for recommendation is trained. Combining product label information of the financial products and information browsed by the user (for example, the information browsed by the user and information similar to the information browsed by the user), determining the financial products which the user is interested in, and combining various characteristics to recommend the financial products for the user; financial products can be accurately recommended to the user; the product label information comprises popularity in a long time period and popularity in the long time period, and further financial products are accurately recommended to users.

Fig. 6 is a schematic structural diagram of an apparatus for recommending financial products according to an embodiment of the present application, and as shown in fig. 6, the apparatus includes:

the first acquiring unit 61 is configured to acquire a to-be-processed financial product list, where the to-be-processed financial product list includes at least one to-be-processed financial product.

A first determining unit 62 for determining product tag information of the financial product to be processed in the list of financial products to be processed.

The second obtaining unit 63 is used for obtaining key information of the information corresponding to the user.

The second determining unit 64 is configured to input the product tag information and the key information of the financial product to be processed in the financial product list to be processed into a preset recommendation model, so as to obtain a financial product list to be recommended, where the financial product list to be recommended includes at least one financial product to be recommended.

A recommending unit 65 for recommending the financial product to be recommended to the user.

For example, the present embodiment may refer to the above method embodiments, and the principle and the technical effect are similar and will not be described again.

Fig. 7 is a schematic structural diagram of another recommendation apparatus for financial products according to an embodiment of the present application, and based on the embodiment shown in fig. 6, as shown in fig. 7, the product tag information includes one or more of the following: the product category, the risk level, the product data information, the popularity in the first time period and the popularity in the second time period; wherein the time length of the first time period is greater than the time length of the second time period.

In one example, if the product tag information includes a popularity in the first time period and a popularity in the second time period, the first determining unit 62 includes:

the first obtaining module 621 is configured to obtain each first product data information of the to-be-processed financial product in the to-be-processed financial product list in the first time period and each second product data information in the second time period, determine a weight corresponding to each first product data information of the to-be-processed financial product in the first time period, and determine a weight corresponding to each second product data information of the to-be-processed financial product in the second time period.

The first determining module 622 is configured to determine the popularity of the financial product to be processed in the first time period according to the first product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of the first product data information.

The second determining module 623 is configured to determine the popularity of the financial product to be processed in the second time period according to the second product data information of the financial product to be processed in the first time period and the weight corresponding to each piece of the second product data information.

In an example, the second obtaining unit 63 is specifically configured to: the information browsed by the user is obtained, and the key information of the information is extracted by adopting a natural language processing mode.

In an example, the apparatus provided in this embodiment further includes:

a third obtaining unit 71, configured to obtain an interest tag of the user before the second determining unit 64 inputs the product tag information and the key information of the financial product to be processed in the financial product list to be processed into a preset recommendation model to obtain the financial product list to be recommended, and according to the determination result, determine the financial product to be processed in the financial product list to be processed, which is matched with the interest tag; wherein, the financial product to be processed matched with the interested label is the financial product input into the preset recommendation model.

In an example, the apparatus provided in this embodiment further includes:

a training unit 72, configured to obtain a financial product list to be trained, where the financial product list to be trained includes at least one financial product to be trained; determining product label information of financial products to be processed in a financial product list to be trained; acquiring information browsed by a user; acquiring user interest tag information; performing feature extraction on product label information of the financial product to be processed and information browsed by a user to obtain a feature vector; training the initial model according to the feature vectors to obtain an initial recommendation model; and performing incremental iterative training on the initial recommendation model according to the product label information and the user interest label information of the financial product to be processed to obtain the recommendation model for obtaining the financial product to be recommended.

In one example, if the product label information includes a popularity in a first time period and a popularity in a second time period, the training unit 71 is specifically configured to, when determining the product label information of the financial product to be processed in the financial product list to be trained:

In one example, the user interest tag information comprises the interest degree of the user in a first time period and the interest degree of the user in a second time period, wherein the time length of the first time period is greater than that of the second time period; when obtaining the user interest label information, the training unit 71 is specifically configured to:

Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 8, the electronic device includes: a memory 81, a processor 82;

a memory 81; a memory for storing instructions executable by the processor 82;

wherein the processor 82 is configured to perform the methods provided in the above embodiments.

The electronic device further comprises a receiver 83 and a transmitter 84. The receiver 83 is used for receiving commands and data sent by an external device, and the transmitter 84 is used for sending commands and data to the external device.

FIG. 9 is a block diagram illustrating an electronic device, which may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like, in accordance with an exemplary embodiment.

The apparatus 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.

The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing components 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

The memory 804 is configured to store various types of data to support operations at the apparatus 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices 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 or optical disks.

Power components 806 provide power to the various components of device 800. The power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the apparatus 800.

The multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed status of the device 800, the relative positioning of the components, such as a display and keypad of the device 800, the sensor assembly 814 may also detect a change in the position of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration/deceleration of the device 800, and a change in the temperature of the device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 816 is configured to facilitate communications between the apparatus 800 and other devices in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, communications component 816 further includes a Near Field Communications (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.

In an exemplary embodiment, a non-transitory computer-readable storage medium comprising instructions, such as the memory 804 comprising instructions, executable by the processor 820 of the device 800 to perform the above-described method is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

A non-transitory computer readable storage medium, in which instructions, when executed by a processor of an electronic device, enable the electronic device to perform the above-described method.

An embodiment of the present application further provides a computer program product, where the computer program product includes: a computer program, stored in a readable storage medium, from which at least one processor of the electronic device can read the computer program, the at least one processor executing the computer program causing the electronic device to perform the solution provided by any of the embodiments described above.

Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

It will be understood that the present application is not limited to the precise arrangements 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 application is limited only by the appended claims.

Claims

1. A method for recommending financial products, said method comprising:

and recommending the financial product to be recommended to the user.

2. The method of claim 1, wherein the product tag information comprises one or more of: the product category, the risk level, the product data information, the popularity in the first time period and the popularity in the second time period; wherein the time length of the first time period is greater than the time length of the second time period.

3. The method of claim 2, wherein determining the product tag information for the financial product to be processed in the list of financial products to be processed if the product tag information includes a popularity of the first time period and a popularity of the second time period comprises:

4. The method of claim 1, wherein obtaining key information of the user corresponding information comprises:

5. The method of claim 1, wherein before inputting the product tag information of the financial product to be processed in the list of financial products to be processed and the key information of the information into a preset recommendation model to obtain the list of financial products to be recommended, the method further comprises:

6. The method according to any one of claims 1-5, further comprising:

7. The method of claim 6, wherein determining the product tag information for the financial product to be processed in the list of financial products to be trained if the product tag information includes a popularity of a first time period and a popularity of a second time period, wherein a length of time of the first time period is greater than a length of time of the second time period comprises:

8. The method of claim 6, wherein the user interest tag information comprises a user interest level in a first time period and a user interest level in a second time period, wherein the time length of the first time period is longer than the time length of the second time period; acquiring user interest tag information, comprising:

9. An apparatus for recommending financial products, said apparatus comprising:

10. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;

the memory stores computer-executable instructions;

the processor executes computer-executable instructions stored by the memory to implement the method of any of claims 1-8.

11. A computer-readable storage medium having computer-executable instructions stored therein, which when executed by a processor, are configured to implement the method of any one of claims 1-8.

12. A computer program product, characterized in that it comprises a computer program which, when being executed by a processor, carries out the method of any one of claims 1-8.