WO2022089190A1

WO2022089190A1 - Product recommendation method and apparatus, and electronic device and readable storage medium

Info

Publication number: WO2022089190A1
Application number: PCT/CN2021/123176
Authority: WO
Inventors: 张�杰
Original assignee: 深圳壹账通智能科技有限公司
Priority date: 2020-11-02
Filing date: 2021-10-12
Publication date: 2022-05-05
Also published as: CN112148995A

Abstract

The present application relates to the technical field of artificial intelligence. Disclosed is a product recommendation method, comprising: determining first features of users, and grouping the users on the basis of the first features, so as to obtain a plurality of user groups; calculating a product permeability array corresponding to each user group in the plurality of user groups; determining a target user group corresponding to a user to be given a recommendation; acquiring a second feature factor corresponding to a product recommendation request, determining, on the basis of the second feature factor and second historical data, a second feature of the user to be given a recommendation, and inputting the second feature into an attribute analysis model, so as to obtain a target attribute value of the user to be given a recommendation; and when the target attribute value is less than a pre-set threshold value, on the basis of the product permeability array corresponding to the target user group, recommending a target product to the user to be given a recommendation. Further provided are a product recommendation apparatus, an electronic device and a readable storage medium. By means of the present application, the success rate of product recommendation is improved.

Description

Product recommendation method, device, electronic device and readable storage medium

This application claims the priority of the Chinese patent application with the application number CN202011206222.0 and the title of "Product Recommendation Method, Apparatus, Electronic Equipment and Readable Storage Medium" filed with the China Patent Office on November 2, 2020, the entire contents of which are Incorporated herein by reference.

technical field

The present application relates to the field of artificial intelligence technology and data analysis, and in particular, to a product recommendation method, device, electronic device, and readable storage medium.

Background technique

With the development of science and technology, the application of product recommendation in people's life is more and more widely, for example, article push, product information recommendation, etc. The inventor realized that at present, it is usually based on the user's own related information, such as historical browsing information, Purchase information for recommendation, so that the success rate of recommendation is not high. Therefore, a product recommendation method is urgently needed to improve the success rate of recommendation.

SUMMARY OF THE INVENTION

The product recommendations provided in this application include:

Obtaining the first historical data of each user in the database, determining the first characteristic of each user based on the first characteristic factor and the first historical data, and grouping each user based on the first characteristic to obtain a plurality of user groups;

Calculate a product penetration rate array corresponding to each user group in the multiple user groups based on the first historical data;

Parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, acquire the second historical data of the user to be recommended from the database based on the identifier, and determine the The target user group corresponding to the user to be recommended;

Obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into the attribute analysis model to obtain the result. Describe the target attribute value of the user to be recommended;

When the target attribute value is less than the preset threshold, a target product is recommended for the user to be recommended based on the product penetration rate array corresponding to the target user group.

The present application also provides a product recommendation device, the device comprising:

The grouping module is used for obtaining the first historical data of each user in the database, determining the first characteristic of each user based on the first characteristic factor and the first historical data, and grouping each user based on the first characteristic to obtain multiple user groups;

a calculation module, configured to calculate the product penetration rate array corresponding to each user group in the multiple user groups based on the first historical data;

The parsing module is configured to parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, and obtain the second historical data of the user to be recommended from the database based on the identifier , and determine the target user group corresponding to the user to be recommended;

A determination module, configured to obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into The attribute analysis model obtains the target attribute value of the user to be recommended;

A recommendation module, configured to recommend a target product for the user to be recommended based on the product penetration rate array corresponding to the target user group when the target attribute value is less than a preset threshold.

The present application also provides an electronic device, the electronic device comprising:

at least one processor; and,

a memory communicatively coupled to the at least one processor; wherein,

The memory stores a product recommendation program executable by the at least one processor, and the product recommendation program is executed by the at least one processor to enable the at least one processor to perform the following steps:

The present application also provides a computer-readable storage medium, where a product recommendation program is stored on the computer-readable storage medium, and the product recommendation program can be executed by one or more processors to implement the following steps:

Description of drawings

1 is a schematic flowchart of a product recommendation method provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of a module of a product recommendation device provided by an embodiment of the present application;

3 is a schematic structural diagram of an electronic device for implementing a product recommendation method provided by an embodiment of the present application;

The realization, functional characteristics and advantages of the purpose of the present application will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

It should be noted that the descriptions involving "first", "second", etc. in this application are only for the purpose of description, and should not be construed as indicating or implying their relative importance or implying the number of indicated technical features . Thus, a feature delimited with "first", "second" may expressly or implicitly include at least one of that feature. In addition, the technical solutions between the various embodiments can be combined with each other, but must be based on the realization by those of ordinary skill in the art. When the combination of technical solutions is contradictory or cannot be realized, it should be considered that the combination of such technical solutions does not exist. , is not within the scope of protection claimed in this application.

The embodiments of the present application may acquire and process related data based on artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. .

The basic technologies of artificial intelligence generally include technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, and mechatronics. Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning/deep learning.

The present application provides a product recommendation method. Referring to FIG. 1 , it is a schematic flowchart of a product recommendation method according to an embodiment of the present application. The method may be performed by an electronic device, which may be implemented by software and/or hardware.

In this embodiment, the product recommendation method includes:

S1. Acquire the first historical data of each user in the database, determine the first characteristic of each user based on the first characteristic factor and the first historical data, and group each user based on the first characteristic to obtain a plurality of user groups .

This embodiment is described by taking a product recommended by a bank customer as an example, the first historical data includes the basic information and asset information of the user, and the first characteristic factor includes the user's gender, age, educational background, age of accounts, holdings in the basic information. Card level, customer level, whether you are a member, whether you have completed risk assessment, whether you are bound to third-party payment, etc. and asset information in time-point AUM (total market value of assets), historical maximum AUM, PPC value (current number of products held), AUM month Daily average value, AUM annual daily average value, etc. The first feature is obtained by splicing the feature values corresponding to each feature in the basic information and the asset information.

The grouping of each user based on the first feature to obtain a plurality of user groups includes:

A1. Use the K-means clustering algorithm to group each user based on the first feature, where K respectively takes a value of each natural number within a preset value range, and a value of K corresponds to a grouping result, and obtains multiple grouping results;

K represents the number of user groups. In this embodiment, K is any natural number from 3 to 10, then it can be divided into 3 user groups, 4 user groups, ..., 9 user groups, Divided into 10 user groups with a total of 8 grouping results.

Take K=3 as an example to illustrate the user grouping process: arbitrarily take the first features of 3 users as the three initial cluster centers, then calculate the distance between the remaining users and the first features of each cluster center, and assign each user to To the cluster closest to it, each time a user is assigned, the cluster center of the cluster will be recalculated according to the existing users in the cluster, and so on until all users are grouped.

A2, determine the center user of each user group corresponding to each kind of grouping result in the multiple grouping results, and calculate the corresponding silhouette coefficient of each kind of grouping result based on the first feature of the center user;

The existing calculation method of the silhouette coefficient has high complexity, wherein the calculation formula of the silhouette coefficient corresponding to the user is:

Among them, S _ij represents the silhouette coefficient corresponding to the j th user in the ith grouping result,

represents the average distance from the first feature of the jth user in the i-th grouping result to the first features of other users in the same user group,

Indicates the minimum value of the average distance between the first feature of the jth user and the first features of other user groups in the ith grouping result.

The calculation formula of the contour coefficient corresponding to the grouping result is:

Among them, S _i represents the contour coefficient corresponding to the i-th grouping result, S _ij represents the contour coefficient corresponding to the j-th user in the i-th grouping result, and n represents the total number of users.

The silhouette coefficient is an evaluation method for the quality of the grouping results, which reflects the cohesion and separation of the clustering method. If the inner clustering of the same cluster is higher, and the separation degree of different clusters is higher, the clustering effect is better, and the closer S _ij is to 1, it means

The smaller the value, the better the clustering effect.

It can be seen from the above formula that when calculating the silhouette coefficient corresponding to each user, the distance from the first eigenvalue of other users should be calculated separately, and finally the distances of all users are averaged, and the time complexity is n ² . In order to reduce the time complexity, the present application proposes a simplified method for calculating the silhouette coefficient. During the clustering process, the centroid of each cluster is known due to the K-means algorithm. The centroid is the core point of each cluster, and other points in the cluster are close to it, so the centroid can approximately represent the entire cluster. When calculating the contour coefficient, only the distance between the centroids is calculated, that is, the formula for calculating the contour coefficient in this embodiment remains unchanged, but the meanings of the parameters are different, and the amount of calculation is greatly reduced.

The determining of the central user (ie the centroid) of each user group corresponding to each of the multiple grouping results includes:

Calculate the average value of the first characteristics of all users in each user group corresponding to each grouping result, and take the user with the smallest absolute value of the difference between the first characteristic and the average value in each user group as the central user of each user group.

In other embodiments, the central user of each user group may also be determined in other manners, and the present application does not limit the determination manner of the central user.

The calculation formula of the contour coefficient in this embodiment is:

Among them, S _pq represents the silhouette coefficient corresponding to the qth user group in the pth grouping result,

represents the average distance from the first feature of the central user of the qth user group in the pth grouping result to the first features of other users in the same user group,

Represents the minimum value of the average distance between the first feature of the central user of the qth user group and the first features of other user groups in the _pth grouping result, Sp represents the contour coefficient corresponding to the pth grouping result, m represents the th The total number of user groups in p types of grouping results.

It can be seen that the present application simplifies the complexity of calculating the silhouette coefficient from n ² to mn, where n is the total number of users and m is the number of user groups. Therefore, the present application greatly improves the grouping efficiency.

A3. The grouping result with the contour coefficient closest to the preset value is used as the target grouping result.

In this embodiment, the preset value is 1, and the grouping result with the contour coefficient closest to 1 is used as the target grouping result.

S2. Calculate, based on the first historical data, a product penetration rate array corresponding to each user group in the multiple user groups.

The calculating, based on the first historical data, an array of product penetration rates corresponding to each of the multiple user groups includes:

B1. Determine the product corresponding to each user in each user group based on the first historical data;

B2. Calculate the proportion of users corresponding to various products in each user group;

B3. Determine the product penetration rate array corresponding to each user group based on the user ratio.

In this embodiment, the first historical data also includes the user's product purchase information, the product penetration rate in the product penetration rate array corresponding to a certain user group is the number of people who purchased each product in the user group and the total users of the user group percentage of the quantity.

For example, the proportion of users corresponding to each product in each user group is shown in Table 1 below:

	活期current	定期regular	理财financial management	基金fund	保险Insurance	……...	信托trust
用户组1User group 1	20％20%	5％5%	50％50%	2％2%	6％6%	……...	2％2%
用户组2User group 2	30％30%	1％1%	1％1%	62％62%	1％1%	……...	1％1%
用户组3User group 3	13％13%	20％20%	1％1%	2％2%	52％52%	……...	2％2%
用户组4User group 4	9％9%	78％78%	2％2%	5％5%	2％2%	……...	1％1%

Table 1

It can be seen from Table 1 above that the product penetration rate array corresponding to user group 1 is {20%, 5%, 50%, 2%, 6%, ..., 2%}. According to the product penetration rate array, different user groups can be analyzed. The degree of preference for different products can be used for product recommendation.

S3. Parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, acquire the second historical data of the user to be recommended from the database based on the identifier, and determine the target user group corresponding to the user to be recommended;

If the historical data of the user to be recommended can be obtained from the database, it means that the user to be recommended is an existing user, and the target user group information corresponding to the user to be recommended can be obtained based on the user ID. The absolute value of the difference between the recommended user and the first characteristic of the central user of each user group, and the user group to which the central user corresponding to the smallest absolute value of the difference belongs is used as the target user group of the user to be recommended.

In this embodiment, the method further includes:

If the second historical data of the user to be recommended cannot be obtained from the database, a preset product list is recommended to the user to be recommended.

If the historical data cannot be obtained, it means that the user to be recommended is a new user and no data information has been generated. At this time, the preconfigured product list for the new user is recommended to the new user.

S4. Obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into an attribute analysis model Obtain the target attribute value of the user to be recommended.

In this embodiment, the attribute analysis model is used to analyze the loss probability of users, and the attribute analysis model is a random forest model.

The second feature factor includes volatility feature, trend feature, product hobby feature, and activity feature. In this embodiment, time slices are used to construct the second feature of the user, for example, the first 6 months of data are selected from the second historical data. The data and the data of the previous 3 months are used to determine the eigenvalues corresponding to each eigenfactor of the user.

The characteristic values corresponding to the volatility characteristics include: the asset range in the past three months (the difference between the maximum asset value and the asset minimum value), the asset standard deviation in the past three months, the asset range in the past six months, and the asset range in the past six months. The standard deviation of assets in 6 months; the characteristic values corresponding to trend characteristics include: asset growth rate in the past 3 months, growth rate of the number of individual gold products held in the past 3 months, asset growth rate in the past 6 months, and recent asset growth rate. The growth rate of the number of individual gold products held in 6 months; the characteristic values corresponding to the characteristics of product hobbies include: the proportion of current accounts, the proportion of regular accounts, the proportion of wealth management, the proportion of funds, the proportion of asset management, and the proportion of insurance; activeness The feature values corresponding to the feature include: ageing, average number of transactions in the past 3 months, average number of transactions in the past 6 months, etc.

The second feature is obtained by splicing the feature values corresponding to the aforementioned volatility feature, trend feature, product hobby feature, and activity feature, and the second feature is input into the attribute analysis model to obtain the target attribute value of the user to be recommended (ie, loss probability) .

S5. When the target attribute value is less than a preset threshold, recommend a target product for the user to be recommended based on the product penetration rate array corresponding to the target user group.

When the target attribute value (that is, the churn probability) is less than the preset threshold, it means that the probability of user churn is low. At this time, the product penetration rate array corresponding to the target user group to which the user to be recommended belongs is obtained, and the products with the highest product penetration rate are used as The target product is recommended to the user to be recommended.

In this embodiment, after inputting the second feature into the attribute analysis model to obtain the target attribute value of the user to be recommended, the method further includes:

If the target attribute value is greater than the preset threshold, calculate the index values of multiple preset indexes based on the second historical data, and when the index value corresponding to a specified index among the multiple preset indexes is greater than the index threshold , recommending a target product for the user to be recommended based on the specified index.

When the target attribute value (that is, the churn probability) is greater than the preset threshold, it indicates that the user churn probability is high, the preset index is a preset index that has a greater impact on the churn rate, and the user churn can be determined based on the preset index Reasons and recommend suitable products. For example, the standard deviation of assets is a preset indicator. When the standard deviation of assets is greater than the indicator threshold, it means that the assets of the users to be recommended have changed greatly. In order to reduce the standard deviation of assets, you can recommend large-denomination certificates of deposit such as Long-term financial products are recommended to users to be recommended.

It can be seen from the above embodiments that, in the product recommendation method proposed by the present application, firstly, each user is grouped according to the first feature to obtain a plurality of user groups, and the product penetration rate array corresponding to each user group is calculated based on the first historical data. The degree of preference of different user groups for different products is analyzed, which can be helpful for accurate recommendation; then, the second historical data of the user to be recommended is obtained, and the target user group corresponding to the user to be recommended is determined. The historical data determines the second feature of the user to be recommended, and the second feature is input into the attribute analysis model to obtain the target attribute value (ie, the churn rate) of the user to be recommended. When the target attribute value is less than the preset threshold, the product corresponding to the target user group The penetration rate array is the target product recommended by the user to be recommended. In this step, product recommendation is performed based on the product penetration rate array when the user churn rate is low, so that the success rate of product recommendation is higher. Therefore, the present application improves the success rate of product recommendation.

As shown in FIG. 2 , it is a schematic diagram of a module of a product recommendation apparatus provided by an embodiment of the present application.

The product recommendation apparatus 100 described in this application may be installed in an electronic device. According to the implemented functions, the product recommendation apparatus 100 may include a grouping module 110 , a calculation module 120 , a parsing module 130 , a determination module 140 and a recommendation module 150 . The modules described in this application may also be referred to as units, which refer to a series of computer program segments that can be executed by the processor of an electronic device and can perform fixed functions, and are stored in the memory of the electronic device.

In this embodiment, the functions of each module/unit are as follows:

The grouping module 110 is configured to obtain the first historical data of each user in the database, determine the first characteristic of each user based on the first characteristic factor and the first historical data, and group each user based on the first characteristic to obtain Multiple user groups.

The smaller the value, the better the clustering effect.

The calculation formula of the contour coefficient in this embodiment is:

The calculation module 120 is configured to calculate a product penetration rate array corresponding to each user group in the plurality of user groups based on the first historical data.

For example, the proportion of users corresponding to each product in each user group is shown in Table 1 above.

The parsing module 130 is configured to parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, and obtain the second history of the user to be recommended from the database based on the identifier data, and determine the target user group corresponding to the user to be recommended;

In this embodiment, the parsing module 130 is further configured to:

The determining module 140 is configured to obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and assign the second characteristic The attribute analysis model is input to obtain the target attribute value of the user to be recommended.

The recommendation module 150 is configured to recommend a target product for the user to be recommended based on the product penetration rate array corresponding to the target user group when the target attribute value is less than a preset threshold.

In this embodiment, after inputting the second feature into the attribute analysis model to obtain the target attribute value of the user to be recommended, the recommendation module 150 is further configured to:

As shown in FIG. 3 , it is a schematic structural diagram of an electronic device for implementing a product recommendation method according to an embodiment of the present application.

The electronic device 1 is a device that can automatically perform numerical calculation and/or information processing according to pre-set or stored instructions. The electronic device 1 may be a computer, a single network server, a server group composed of multiple network servers, or a cloud based on cloud computing composed of a large number of hosts or network servers, wherein cloud computing is a kind of distributed computing, A super virtual computer consisting of a collection of loosely coupled computers.

In this embodiment, the electronic device 1 includes, but is not limited to, a memory 11 , a processor 12 , and a network interface 13 that can be communicatively connected to each other through a system bus, the memory 11 stores a product recommendation program 10 , and the product recommendation program 10 is executable by the processor 12 . FIG. 3 only shows the electronic device 1 having the components 11-13 and the product recommendation program 10. Those skilled in the art can understand that the structure shown in FIG. 3 does not constitute a limitation on the electronic device 1, and may include a Fewer or more components are shown, or some components are combined, or a different arrangement of components.

The memory 11 includes a memory and at least one type of readable storage medium. The memory provides a cache for the operation of the electronic device 1; the readable storage medium can be, for example, flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM) ), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. non-volatile storage media. The readable storage medium may also be a volatile storage medium. In some embodiments, the readable storage medium may be an internal storage unit of the electronic device 1, such as a hard disk of the electronic device 1; in other embodiments, the storage medium may also be an external storage device of the electronic device 1, such as A pluggable hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card (Flash Card), etc., are equipped on the electronic device 1. In this embodiment, the readable storage medium of the memory 11 is generally used to store the operating system and various application software installed in the electronic device 1 , for example, to store the code of the product recommendation program 10 in an embodiment of the present application. In addition, the memory 11 can also be used to temporarily store various types of data that have been output or will be output.

In some embodiments, the processor 12 may be a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 12 is generally used to control the overall operation of the electronic device 1, such as performing control and processing related to data interaction or communication with other devices. In this embodiment, the processor 12 is configured to run the program code or process data stored in the memory 11, for example, run the product recommendation program 10 and the like.

The network interface 13 may include a wireless network interface or a wired network interface, and the network interface 13 is used to establish a communication connection between the electronic device 1 and a client (not shown in the figure).

Optionally, the electronic device 1 may further include a user interface, and the user interface may include a display (Display), an input unit such as a keyboard (Keyboard), and an optional user interface may also include a standard wired interface and a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, and the like. The display may also be appropriately called a display screen or a display unit, which is used for displaying information processed in the electronic device 1 and for displaying a visualized user interface.

It should be understood that the embodiments are only used for illustration, and are not limited by this structure in the scope of the patent application.

The product recommendation program 10 stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, and when running in the processor 12, can realize:

Specifically, for the specific implementation method of the above-mentioned product recommendation program 10 by the processor 12, reference may be made to the description of the relevant steps in the corresponding embodiment of FIG. 1 , which will not be repeated here. It should be emphasized that, in order to further ensure the privacy and security of the above-mentioned first and second historical data, the above-mentioned first and second historical data may also be stored in a node of a blockchain.

Further, if the modules/units integrated in the electronic device 1 are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. The computer-readable storage medium may be non-volatile or non-volatile. The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only Memory) ).

A product recommendation program 10 is stored on the computer-readable storage medium, and the product recommendation program 10 can be executed by one or more processors to realize the following steps:

In the several embodiments provided in this application, it should be understood that the disclosed apparatus, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there may be other division manners in actual implementation.

The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional module in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of hardware plus software function modules.

It will be apparent to those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

Accordingly, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of the application is to be defined by the appended claims rather than the foregoing description, which is therefore intended to fall within the scope of the claims. All changes within the meaning and scope of the equivalents of , are included in this application. Any reference signs in the claims shall not be construed as limiting the involved claim.

The blockchain referred to in this application is a new application mode of computer technologies such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information to verify its Validity of information (anti-counterfeiting) and generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

Furthermore, it is clear that the word "comprising" does not exclude other units or steps and the singular does not exclude the plural. A plurality of units or means recited in the system claims can also be implemented by one unit or means by software or hardware. Second-class terms are used to denote names and do not denote any particular order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application rather than limitations. Although the present application has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present application can be Modifications or equivalent substitutions can be made without departing from the spirit and scope of the technical solutions of the present application.

Claims

A product recommendation method, wherein the method includes:

Obtaining the first historical data of each user in the database, determining the first characteristic of each user based on the first characteristic factor and the first historical data, and grouping each user based on the first characteristic to obtain a plurality of user groups;

Calculate a product penetration rate array corresponding to each user group in the multiple user groups based on the first historical data;

Parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, acquire the second historical data of the user to be recommended from the database based on the identifier, and determine the The target user group corresponding to the user to be recommended;

Obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into the attribute analysis model to obtain the result. Describe the target attribute value of the user to be recommended;

When the target attribute value is less than the preset threshold, a target product is recommended for the user to be recommended based on the product penetration rate array corresponding to the target user group.
The product recommendation method according to claim 1, wherein the grouping of each user based on the first characteristic to obtain a plurality of user groups comprises:

Based on the first feature, the K-means clustering algorithm is used to group each user, wherein K respectively takes the value of each natural number within a preset value range, and one value of K corresponds to one kind of grouping result, and various grouping results are obtained. ;

Determine the center user of each user group corresponding to each kind of grouping result in the multiple grouping results, and calculate the silhouette coefficient corresponding to each kind of grouping result based on the first feature of the center user;

The grouping result with the silhouette coefficient closest to the preset value is taken as the target grouping result.
The product recommendation method according to claim 2, wherein said determining the central user of each user group corresponding to each of the plurality of grouping results comprises:

Calculate the average value of the first characteristics of all users in each user group corresponding to each grouping result, and take the user with the smallest absolute value of the difference between the first characteristic and the average value in each user group as the central user of each user group.
The product recommendation method according to claim 2, wherein the calculation formula of the silhouette coefficient is:

Among them, S pq represents the silhouette coefficient corresponding to the qth user group in the pth grouping result,
represents the average distance from the first feature of the central user of the qth user group in the pth grouping result to the first features of other users in the same user group,
Represents the minimum value of the average distance between the first feature of the central user of the qth user group and the first features of other user groups in the pth grouping result, Sp represents the contour coefficient corresponding to the pth grouping result, m represents the th The total number of user groups in p types of grouping results.
The product recommendation method according to claim 1, wherein after inputting the second feature into an attribute analysis model to obtain the target attribute value of the user to be recommended, the method further comprises:

If the target attribute value is greater than the preset threshold, calculate the index values of multiple preset indexes based on the second historical data, and when the index value corresponding to a specified index among the multiple preset indexes is greater than the index threshold , recommending a target product for the user to be recommended based on the specified index.
The product recommendation method according to claim 1, wherein calculating the product penetration rate array corresponding to each user group in the plurality of user groups based on the first historical data comprises:

Determine the product corresponding to each user in each user group based on the first historical data;

Calculate the proportion of users corresponding to various products in each user group;

The product penetration rate array corresponding to each user group is determined based on the user ratio.
The product recommendation method according to claim 1, wherein after acquiring the second historical data of the user to be recommended from the database based on the identifier, the method further comprises:

If the second historical data of the user to be recommended cannot be obtained from the database, a preset product list is recommended to the user to be recommended.
A product recommendation device, wherein the device includes:

The grouping module is used for obtaining the first historical data of each user in the database, determining the first characteristic of each user based on the first characteristic factor and the first historical data, and grouping each user based on the first characteristic to obtain multiple user groups;

a calculation module, configured to calculate the product penetration rate array corresponding to each user group in the multiple user groups based on the first historical data;

The parsing module is configured to parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, and obtain the second historical data of the user to be recommended from the database based on the identifier , and determine the target user group corresponding to the user to be recommended;

A determination module, configured to obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into The attribute analysis model obtains the target attribute value of the user to be recommended;

A recommendation module, configured to recommend a target product for the user to be recommended based on the product penetration rate array corresponding to the target user group when the target attribute value is less than a preset threshold.
An electronic device, wherein the electronic device comprises:

at least one processor; and,

a memory communicatively coupled to the at least one processor; wherein,

The memory stores a product recommendation program executable by the at least one processor, and the product recommendation program is executed by the at least one processor to enable the at least one processor to perform the following steps:

Obtaining the first historical data of each user in the database, determining the first characteristic of each user based on the first characteristic factor and the first historical data, and grouping each user based on the first characteristic to obtain a plurality of user groups;

Calculate a product penetration rate array corresponding to each user group in the multiple user groups based on the first historical data;

Parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, acquire the second historical data of the user to be recommended from the database based on the identifier, and determine the The target user group corresponding to the user to be recommended;

Obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into the attribute analysis model to obtain the result. Describe the target attribute value of the user to be recommended;

When the target attribute value is less than the preset threshold, a target product is recommended for the user to be recommended based on the product penetration rate array corresponding to the target user group.
The electronic device according to claim 9, wherein the grouping of each user based on the first characteristic to obtain a plurality of user groups comprises:

Based on the first feature, the K-means clustering algorithm is used to group each user, wherein K respectively takes the value of each natural number within a preset value range, and one value of K corresponds to one kind of grouping result, and various grouping results are obtained. ;

Determine the center user of each user group corresponding to each kind of grouping result in the multiple grouping results, and calculate the silhouette coefficient corresponding to each kind of grouping result based on the first feature of the center user;

The grouping result with the silhouette coefficient closest to the preset value is taken as the target grouping result.
The electronic device according to claim 10, wherein the determining the central user of each user group corresponding to each of the plurality of grouping results comprises:

Calculate the average value of the first characteristics of all users in each user group corresponding to each grouping result, and take the user with the smallest absolute value of the difference between the first characteristic and the average value in each user group as the central user of each user group.
The electronic device according to claim 10, wherein the calculation formula of the silhouette coefficient is:

Among them, S pq represents the silhouette coefficient corresponding to the qth user group in the pth grouping result,
represents the average distance from the first feature of the central user of the qth user group in the pth grouping result to the first features of other users in the same user group,
Represents the minimum value of the average distance between the first feature of the central user of the qth user group and the first features of other user groups in the pth grouping result, Sp represents the contour coefficient corresponding to the pth grouping result, m represents the th The total number of user groups in p types of grouping results.
The electronic device according to claim 9, wherein after the second feature is input into the attribute analysis model to obtain the target attribute value of the user to be recommended, the product recommendation program further implements the following steps when executed by the processor:

If the target attribute value is greater than the preset threshold, calculate the index values of multiple preset indexes based on the second historical data, and when the index value corresponding to a specified index among the multiple preset indexes is greater than the index threshold , recommending a target product for the user to be recommended based on the specified index.
The electronic device according to claim 9, wherein calculating the product penetration rate array corresponding to each user group in the plurality of user groups based on the first historical data comprises:

Determine the product corresponding to each user in each user group based on the first historical data;

Calculate the proportion of users corresponding to various products in each user group;

The product penetration rate array corresponding to each user group is determined based on the user ratio.
The electronic device according to claim 9, wherein after acquiring the second historical data of the user to be recommended from the database based on the identifier, the product recommendation program further implements the following steps when executed by the processor:

If the second historical data of the user to be recommended cannot be obtained from the database, a preset product list is recommended to the user to be recommended.
A computer-readable storage medium, wherein a product recommendation program is stored on the computer-readable storage medium, and the product recommendation program can be executed by one or more processors to realize the following steps:

Obtaining the first historical data of each user in the database, determining the first characteristic of each user based on the first characteristic factor and the first historical data, and grouping each user based on the first characteristic to obtain a plurality of user groups;

Calculate a product penetration rate array corresponding to each user group in the multiple user groups based on the first historical data;

Parse the product recommendation request sent by the user based on the client, obtain the identifier of the user to be recommended carried in the product recommendation request, acquire the second historical data of the user to be recommended from the database based on the identifier, and determine the The target user group corresponding to the user to be recommended;

Obtain the second characteristic factor corresponding to the product recommendation request, determine the second characteristic of the user to be recommended based on the second characteristic factor and the second historical data, and input the second characteristic into the attribute analysis model to obtain the result. Describe the target attribute value of the user to be recommended;

When the target attribute value is less than the preset threshold, a target product is recommended for the user to be recommended based on the product penetration rate array corresponding to the target user group.
The computer-readable storage medium of claim 16, wherein the grouping of the respective users based on the first characteristic to obtain a plurality of user groups comprises:

Based on the first feature, the K-means clustering algorithm is used to group each user, wherein K respectively takes the value of each natural number within a preset value range, and one value of K corresponds to one kind of grouping result, and various grouping results are obtained. ;

Determine the center user of each user group corresponding to each kind of grouping result in the multiple grouping results, and calculate the silhouette coefficient corresponding to each kind of grouping result based on the first feature of the center user;

The grouping result with the silhouette coefficient closest to the preset value is taken as the target grouping result.
The computer-readable storage medium of claim 17, wherein the central user of each user group corresponding to each of the plurality of grouping results is determined comprising:

Calculate the average value of the first characteristics of all users in each user group corresponding to each grouping result, and take the user with the smallest absolute value of the difference between the first characteristic and the average value in each user group as the central user of each user group.
The computer-readable storage medium of claim 17, wherein the calculation formula of the silhouette coefficient is:

Among them, S pq represents the silhouette coefficient corresponding to the qth user group in the pth grouping result,
represents the average distance from the first feature of the central user of the qth user group in the pth grouping result to the first features of other users in the same user group,
Represents the minimum value of the average distance between the first feature of the central user of the qth user group and the first features of other user groups in the pth grouping result, Sp represents the contour coefficient corresponding to the pth grouping result, m represents the th The total number of user groups in p types of grouping results.
The computer-readable storage medium according to claim 16, wherein after the second feature is input into the attribute analysis model to obtain the target attribute value of the user to be recommended, the product recommendation program further implements when the processor is executed. Follow the steps below:

If the target attribute value is greater than the preset threshold, calculate the index values of multiple preset indexes based on the second historical data, and when the index value corresponding to a specified index among the multiple preset indexes is greater than the index threshold , recommending a target product for the user to be recommended based on the specified index.