CN114549122A

CN114549122A - Model training method, commodity recommendation device, equipment and storage medium

Info

Publication number: CN114549122A
Application number: CN202210130576.4A
Authority: CN
Inventors: 文豪
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-02-11
Filing date: 2022-02-11
Publication date: 2022-05-27

Abstract

The disclosure relates to the technical field of artificial intelligence, in particular to the technical field of model training, and can be applied to a commodity recommendation scene. The specific implementation scheme is as follows: monitoring the operating time of the shopping platform; when the running time reaches one training node of the current training period, calling a first commodity recommendation model corresponding to a first historical training node before the current training period, wherein the first commodity recommendation model is obtained by training based on at least part of first sample data before the first historical training node; acquiring second sample data from the first historical training node to the training node in a time period; and training the first commodity recommendation model based on the second sample data to obtain a target commodity recommendation model. The method can ensure that the fluctuation of the recommendation result output by the trained target commodity recommendation model is small, and the accuracy of the recommendation result is ensured.

Description

Model training method, commodity recommendation device, equipment and storage medium

Technical Field

The disclosure relates to the technical field of artificial intelligence, in particular to the technical field of model training, and can be applied to a scene of commodity recommendation.

Background

In some e-commerce platforms, a recommendation model is used for recommending corresponding brought resources for a user, wherein the recommendation model is trained based on purchasing behavior data of the user. However, in the starting stage of the e-commerce platform, the purchasing behavior data of the user of the e-commerce platform is less, which causes the inaccurate recommendation result output by the trained recommendation model and affects the business income of the e-commerce platform.

Disclosure of Invention

The disclosure provides a model training method, a commodity recommendation device, equipment and a storage medium.

According to a first aspect of the present disclosure, there is provided a training method of a commodity recommendation model, including:

monitoring the operating time of the shopping platform;

when the running time reaches one training node of the current training period, calling a first commodity recommendation model corresponding to a first historical training node before the current training period, wherein the first commodity recommendation model is obtained by training based on at least part of first sample data before the first historical training node;

acquiring second sample data from the first historical training node to the training node in a time period;

and training the first commodity recommendation model based on the second sample data to obtain a target commodity recommendation model.

According to a second aspect of the present disclosure, there is provided a commodity recommendation method including:

acquiring characteristic data of a target user in a shopping platform;

inputting the characteristic data into a target commodity recommendation model, and inputting a commodity recommendation result aiming at a target user through the target commodity recommendation model, wherein the target commodity recommendation model is obtained based on the training method as claimed in any one of claims 1 to 6;

and pushing a commodity recommendation result.

According to a third aspect of the present disclosure, there is provided a training apparatus for a commodity recommendation model, including:

the time monitoring module is used for monitoring the operation time of the shopping platform;

the model calling module is used for calling a first commodity recommendation model corresponding to a first historical training node before the current training period when the running time is determined to reach one training node of the current training period, wherein the first commodity recommendation model is obtained by training based on at least part of first sample data before the first historical training node;

the sample acquisition module is used for acquiring second sample data in a time period from the first historical training node to the training node;

and the target model training module is used for training the first commodity recommendation model based on the second sample data to obtain a target commodity recommendation model.

According to a fourth aspect of the present disclosure, there is provided an article recommendation device including:

the characteristic acquisition module is used for acquiring characteristic data of a target user in a shopping platform;

the result output module is used for inputting the characteristic data into a target commodity recommendation model, inputting a commodity recommendation result aiming at a target user through the target commodity recommendation model, and obtaining the target commodity recommendation model based on the training method provided by the first aspect;

and the result pushing module is used for pushing the commodity recommendation result.

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

at least one processor; and a memory communicatively coupled to the at least one processor;

the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the training method of the commodity recommendation model provided by the first aspect or the commodity recommendation method provided by the second aspect.

According to a sixth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the training method of the product recommendation model provided in the first aspect or the product recommendation method provided in the second aspect.

According to a seventh aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the training method of the merchandise recommendation model provided in the first aspect or the merchandise recommendation method provided in the second aspect.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

The technical scheme provided by the disclosure has the following beneficial effects:

in the technical scheme of the disclosure, the sample data used by the training model contains a large amount of historical sample data in a longer time period before the current training period, so that the fluctuation of the recommendation result output by the trained target commodity recommendation model is ensured to be smaller; the sample data used by the training model also comprises sample data in the current training period, so that the recommendation result output by the trained target commodity recommendation model is ensured to better conform to the current use scene, and the accuracy of the recommendation result can be ensured under the condition of the sample data of the cold start user and the cold start resource.

Drawings

The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

fig. 1 is a flowchart illustrating a training method of a commodity recommendation model according to an embodiment of the present disclosure;

FIG. 2 is a flowchart illustrating a training method for a first commodity recommendation model according to an embodiment of the present disclosure;

fig. 3 is a flowchart illustrating a product recommendation method provided by an embodiment of the present disclosure;

FIG. 4 is a schematic diagram illustrating a training apparatus for a merchandise recommendation model according to an embodiment of the present disclosure;

FIG. 5 is a schematic diagram illustrating a training apparatus for a merchandise recommendation model provided by an embodiment of the disclosure;

FIG. 6 is a schematic diagram illustrating an article recommendation device provided by an embodiment of the present disclosure;

fig. 7 illustrates a schematic block diagram of an example electronic device that may be used to implement the training method of the goods recommendation model or the goods recommendation method provided by the embodiments of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

The embodiment of the disclosure provides a model training method, a commodity recommendation device, equipment and a storage medium, and aims to solve at least one of the above technical problems in the prior art.

Fig. 1 shows a schematic flowchart of a training method for a commodity recommendation model according to an embodiment of the present disclosure, and as shown in fig. 1, the method mainly includes the following steps:

s110: the operating time of the shopping platform is monitored.

In the embodiment of the disclosure, after the shopping platform is released online, the operation time of the shopping platform can be monitored in real time, and it can be understood that the user logs in the shopping platform on the terminal device of the user and online shopping is carried out through the shopping platform. The shopping platform is provided with a corresponding commodity recommendation model, the commodity recommendation model can generate a corresponding commodity recommendation result based on the characteristic data of the user in the shopping platform, and the commodity recommendation result is pushed to the terminal equipment of the user, so that the terminal equipment can display the commodity recommendation result to the user.

S120: when the running time reaches one training node of the current training period, calling a first commodity recommendation model corresponding to a first historical training node before the current training period.

The embodiment of the disclosure can use one unit time length as a training period of a commodity recommendation model of a shopping platform, each training period comprises a plurality of training nodes, and the distance between adjacent training nodes is preset time length. The duration of the training period and the preset duration of the distance between adjacent training nodes may be determined according to actual needs, for example, the training period may be one day (i.e., 24 hours), and the preset duration may be 1 hour. It should be noted that, each time the operating time of the shopping platform reaches one training node, the commodity recommendation model of the shopping platform can be trained once, and the trained commodity recommendation model is released to the line, so that the commodity recommendation result for the user can be predicted based on the newly trained commodity recommendation model.

As described above, a new product recommendation model can be trained every time the shopping platform reaches a training node, that is, each training node may correspond to a version of the product recommendation model, so that the shopping platform has multiple versions of the product recommendation model.

In step S120, a designated training node before the current training cycle may be selected, the designated training node may be defined as a first historical training node, and a product recommendation model corresponding to the first historical training node may be defined as a first product recommendation model. It is understood that as the running time of the shopping platform increases, the number of times that the user uses the shopping platform also increases, so that more sample data is generated. Here, the first commodity recommendation model is trained based at least in part on the first sample data.

S130: and acquiring second sample data from the first historical training node to the training node in a time period.

In the embodiment of the disclosure, for convenience of understanding and expression, sample data generated in a time period from the first historical training node to the training node is defined as second sample data, and the second sample data is used for training the first commodity recommendation model.

Optionally, the first historical training node may be determined according to the number of sample data generated by the shopping platform in a training period and the speed of processing the sample data in the process of training the commodity recommendation model. The first historical training node should meet the following requirements: the total amount of second sample data in a time period from the first historical training node to the training node is smaller than the total amount of sample data used in the model training process in the first preset time. It can be understood that, under the condition that the first historical training node meets the above condition, it can be ensured that the model training process within the first preset time duration (e.g., within 1 hour) can completely process the second sample data within the time period from the first historical training node to the training node, thereby completing the first commodity recommendation model, so that a new commodity recommendation model can be issued on line in time. Here, the first historical training node may be a start time of an nth training period before the current training period, where N is an integer not less than 1. It is understood that the second sample data includes the historical sample data from the first historical training node to the time period from the starting time of the current training cycle and the sample data from the starting time of the current training cycle to the time period of the current training node.

S140: and training the first commodity recommendation model based on the second sample data to obtain a target commodity recommendation model.

According to the training method of the commodity recommendation model provided by the embodiment of the disclosure, the used sample data comprises a large amount of historical sample data in a longer time period before the current training period, so that the fluctuation of the recommendation result output by the trained target commodity recommendation model is ensured to be smaller; the used sample data also comprises sample data in the current training period, so that the recommendation result output by the trained target commodity recommendation model is ensured to be more consistent with the current use scene, and the accuracy of the recommendation result can be ensured under the condition of the sample data of the cold start user and the cold start resource.

Fig. 2 is a flowchart illustrating a training method for a first commodity recommendation model according to an embodiment of the present disclosure, and as shown in fig. 2, the method may mainly include the following steps:

s210: and calling a basic commodity recommendation model.

Here, the basic commodity recommendation model is a recommendation model obtained by training at a first historical training node, and a model training process for obtaining the basic commodity recommendation model is consistent with a model training process for obtaining the target commodity recommendation model, where the basic commodity recommendation model is obtained by training based on first sample data before the first historical training node. It should be noted that the first historical training node may be a starting time of an nth training period before the current training period, where N is an integer not less than 1.

S220: first sub-sample data in an (N +1) th training period before a current training period is acquired.

Taking N as 2 as an example, the first historical training node is the starting time of the 2 nd training period before the current training period, and in step S220, sample data in the 3 rd training period before the current training period may be acquired as the first sub-sample data.

S230: and training the basic commodity recommendation model based on the first sub-sample data to obtain a first commodity recommendation model.

It should be noted that the principle of the specific training process of the basic commodity recommendation model is the same as that of the training process of the target commodity recommendation model, and is not described herein again. In an embodiment of the disclosure, the first sample data and the second sample data each contain purchasing behavior data of the user in the shopping platform.

Fig. 3 shows a schematic flow chart of a product recommendation method provided by the embodiment of the present disclosure, and as shown in fig. 3, the method mainly includes the following steps:

s310: and acquiring characteristic data of the target user in the shopping platform.

Here, the feature data contains purchase behavior data of the user.

S320: inputting the characteristic data into a target commodity recommendation model, and inputting a commodity recommendation result aiming at a target user through the target commodity recommendation model.

The target commodity recommendation model is obtained based on the model training method in the above step S110 to step S14, and the target commodity recommendation model may input a commodity recommendation result for the target user based on the feature data. Specifically, the target commodity recommendation model determines the score of each candidate commodity based on the purchasing behavior data; and screening out the target commodities finally recommended to the user based on the scores, and generating a commodity recommendation result containing the information of the target commodities. Of course, the display order of each candidate product may also be determined based on the scores, and a product recommendation result including the display order of the candidate products may be generated.

S330: and pushing a commodity recommendation result.

Optionally, the target product recommendation model may push the product recommendation result to a terminal device of the user, so that the terminal device may display the product recommendation result to the user.

Based on the same principle as the training method of the commodity recommendation model, the embodiment of the present disclosure provides a training device of a commodity recommendation model, fig. 4 shows a schematic diagram of a training device of a commodity recommendation model provided by the embodiment of the present disclosure, and fig. 5 shows a schematic diagram of a training device of another commodity recommendation model provided by the embodiment of the present disclosure. As shown in fig. 4, the training apparatus 400 for a commodity recommendation model includes a time monitoring module 410, a model retrieving module 420, a sample obtaining module 430, and a target model training module 440.

The time monitoring module 410 is used to monitor the runtime of the shopping platform.

The model retrieving module 420 is configured to, when it is determined that the operation time reaches one training node of the current training period, retrieve a first commodity recommendation model corresponding to a first historical training node before the current training period. The first commodity recommendation model is trained on at least part of first sample data before the first historical training node.

The sample obtaining module 430 is configured to obtain second sample data in a time period from the first historical training node to the training node.

And the target model training module 440 is configured to train the first commodity recommendation model based on the second sample data to obtain a target commodity recommendation model.

In the embodiment of the present disclosure, the current training period includes a plurality of training nodes, and a preset time length is left between adjacent training nodes.

In an embodiment of the present disclosure, the first historical training node is a starting time of an nth training period before the current training period, where N is an integer not less than 1.

In the embodiment of the present disclosure, the total amount of the second sample data in the time period from the first historical training node to the training node is less than the total amount of the sample data used in the model training process in the first preset time period.

In the embodiment of the present disclosure, as shown in fig. 5, the training apparatus 400 of the merchandise recommendation model further includes a first model training module 450, where the first model training module 450 is configured to:

calling a basic commodity recommendation model, wherein the basic commodity recommendation model is obtained by training based on first sample data before a first historical training node;

acquiring first sub-sample data in an (N +1) th training period before a current training period;

and training the basic commodity recommendation model based on the first sub-sample data to obtain a first commodity recommendation model.

In an embodiment of the disclosure, the first sample data and the second sample data each contain purchasing behavior data of the user in the shopping platform.

It can be understood that, in the embodiment of the present disclosure, each module of the training apparatus for a commodity recommendation model has a function of implementing a corresponding step of the training method for a commodity recommendation model. The function can be realized by hardware, and can also be realized by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the functions described above. The modules may be software and/or hardware, and each module may be implemented separately or implemented by integrating a plurality of modules. For the functional description of each module of the training device of the commodity recommendation model, reference may be made to the corresponding description of the training method of the commodity recommendation model, which is not described herein again.

Based on the same principle as the above-described commodity recommendation method, the present disclosure provides a commodity recommendation device, and fig. 6 shows a schematic diagram of a commodity recommendation device provided by the present disclosure. As shown in fig. 6, the product recommendation apparatus 600 includes a feature acquisition module 610, a result output module 620 and a result pushing module 630.

The feature obtaining module 610 is configured to obtain feature data of a target user in a shopping platform.

The result output module 620 inputs the feature data into the target commodity recommendation model, and inputs the commodity recommendation result for the target user through the target commodity recommendation model, where the target commodity recommendation model is obtained by the model training method.

The result pushing module 630 is used for pushing the commodity recommendation result.

In the disclosed embodiment, the characteristic data contains purchasing behavior data of the user.

It can be understood that each module of the product recommendation device in the embodiment of the present disclosure has a function of implementing a corresponding step of the product recommendation method. The function can be realized by hardware, and can also be realized by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the functions described above. The modules can be software and/or hardware, and each module can be implemented independently or by integrating a plurality of modules. For the functional description of each module of the commodity recommendation device, reference may be made to the corresponding description of the commodity recommendation method, which is not described herein again.

In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the personal information of the related user are all in accordance with the regulations of related laws and regulations and do not violate the good customs of the public order.

The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.

FIG. 7 illustrates a schematic block diagram of an example electronic device that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 7, the electronic device 700 includes a computing unit 701, which may perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM)702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the electronic device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

A number of components in the electronic device 700 are connected to the I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, or the like; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the electronic device 700 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.

Computing unit 701 may be a variety of general purpose and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 701 executes the respective methods and processes described above, such as a training method of a commodity recommendation model or a commodity recommendation method. For example, in some embodiments, the training method of the good recommendation model or the good recommendation method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 700 via the ROM 702 and/or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the training method of the commodity recommendation model or the commodity recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured by any other suitable means (e.g., by means of firmware) to perform a training method of a goods recommendation model or a goods recommendation method.

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel or sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.

The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A training method of a commodity recommendation model comprises the following steps:

monitoring the operating time of the shopping platform;

acquiring second sample data from the first historical training node to the training node within a time period;

2. The method of claim 1, wherein the current training period comprises a plurality of training nodes, and adjacent training nodes are separated by a preset time.

3. The method of claim 2, the first historical training node being a starting time of an nth training period prior to the current training period, where N is an integer no less than 1.

4. The method of any of claims 1-3, wherein a total amount of second sample data in a time period from the first historical training node to the training node is less than a total amount of sample data used by a model training process in the first preset duration.

5. The method of claim 3, wherein the first merchandise recommendation model is trained by:

calling a basic commodity recommendation model, wherein the basic commodity recommendation model is obtained by training based on first sample data before the first historical training node;

acquiring first sub-sample data in an (N +1) th training period before the current training period;

and training the basic commodity recommendation model based on the first sub-sample data to obtain the first commodity recommendation model.

6. The method of claim 1, the first sample data and the second sample data each including purchasing behavior data of a user in the shopping platform.

7. A method of merchandise recommendation, comprising:

acquiring characteristic data of a target user in a shopping platform;

inputting the characteristic data into a target commodity recommendation model, and inputting a commodity recommendation result for the target user through the target commodity recommendation model, wherein the target commodity recommendation model is obtained based on the training method of any one of claims 1 to 6;

and pushing the commodity recommendation result.

8. The method of claim 7, the characteristic data comprising purchasing behavior data of a user.

9. A training device for a commodity recommendation model comprises:

the model retrieving module is used for retrieving a first commodity recommendation model corresponding to a first historical training node before the current training period when the running time is determined to reach one training node of the current training period, wherein the first commodity recommendation model is obtained by training based on at least part of first sample data before the first historical training node;

10. The apparatus of claim 9, the current training period comprising a plurality of training nodes, adjacent training nodes being separated by a preset time period.

11. The apparatus of claim 10, the first historical training node being a starting time of an nth training period prior to the current training period, wherein N is an integer no less than 1.

12. The apparatus of any one of claims 911, a total amount of second sample data over a time period from the first historical training node to the training node being less than a total amount of sample data used by a model training process over the first preset duration.

13. The apparatus of claim 11, further comprising a first model training module to:

14. The device of claim 9, the first sample data and the second sample data both including user purchasing behavior data in the shopping platform.

15. An article recommendation device comprising:

a result output module, which inputs the characteristic data into a target commodity recommendation model, and inputs a commodity recommendation result for the target user through the target commodity recommendation model, wherein the target commodity recommendation model is obtained based on the training method of any one of claims 1 to 6;

16. The device of claim 15, the characteristic data comprising purchasing behavior data of a user.

17. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6 or the method of any one of claims 7-8.

18. A non-transitory computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method of any one of claims 1-6 or the method of any one of claims 7-8.

19. A computer program product comprising a computer program which, when executed by a processor, performs the method of any one of claims 1 to 6, or the method of any one of claims 7 to 8.