Unlock instant, AI-driven research and patent intelligence for your innovation.

Recommended commodity determination method and device, electronic equipment and readable storage medium

A product and vector representation technology, applied in the field of neural networks, can solve problems that hinder the performance of recommendation models, negative transfer, and the inability to recognize complex relationships between samples

Pending Publication Date: 2022-06-21
BEIJING SANKUAI ONLINE TECH CO LTD
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Embodiments of the present invention provide a method and device for determining recommended products, electronic equipment, and a readable storage medium, to at least solve the problem of inability to identify complex relationships between samples during joint training of multi-service samples in the related art, thereby causing negative transfer , the technical problems that hinder the further improvement of the performance of the recommendation model

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Recommended commodity determination method and device, electronic equipment and readable storage medium
  • Recommended commodity determination method and device, electronic equipment and readable storage medium
  • Recommended commodity determination method and device, electronic equipment and readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0021] In a specific application scenario, the technical solution in this embodiment recommends commodities for a specific user, obtains user data of the target user, and inputs the user data and commodity data in the database into the recommendation model to determine the target user's Product preference, so as to determine the recalled product corresponding to the target user.

[0022] According to an embodiment of the present invention, a method for determining a recommended product is provided, such as figure 1 As shown, the method may specifically include:

[0023] S102: Determine the first vector representation corresponding to the user data of the target user by the first embedding module in the recommendation model, and determine the second vector representation corresponding to the commodity data by the second embedding module in the recommendation model, wherein the recommendation model is It is pre-trained by preset training samples;

[0024] In this embodiment, t...

Embodiment 2

[0089] According to an embodiment of the present invention, there is also provided a recommended commodity determination device for implementing the above recommended commodity determination method, such as Figure 4 As shown, the device includes:

[0090] 1) The first determination module 40 is used to determine the first vector representation corresponding to the user data of the target user through the first embedding module in the recommendation model, and determine the corresponding product data through the second embedding module in the recommendation model. The second vector representation, wherein the recommendation model is pre-trained by preset training samples;

[0091] 2) A second determination module 42, configured to determine the user vector representation corresponding to the first vector representation through the first mixed gating module of the recommendation model, and, through the second mixed gating module of the recommendation model , determine the comm...

Embodiment 3

[0100] According to an embodiment of the present invention, an electronic device is also provided, including a processor, a memory, and a program or instruction stored on the memory and executable on the processor, the program or instruction being executed by the processor When executed, it implements the steps of the method for determining the recommended product as described above.

[0101] Optionally, in this embodiment, the memory is configured to store program codes for performing the following steps:

[0102] S1, the first vector representation corresponding to the user data of the target user is determined by the first embedding module in the recommendation model, and the second vector representation corresponding to the commodity data is determined by the second embedding module in the recommendation model, wherein the The above recommendation model is pre-trained by preset training samples;

[0103] S2: Determine the user vector representation corresponding to the fi...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a recommended commodity determination method and device, electronic equipment and a readable storage medium. The method comprises the steps that a first vector representation corresponding to user data of a target user is determined through a first embedded module in a recommendation model, a second vector representation corresponding to commodity data is determined through a second embedded module in the recommendation model, and the recommendation model is pre-trained by a preset training sample; determining a user vector representation corresponding to the first vector representation through a first hybrid gating module of the recommendation model, and determining a commodity vector representation corresponding to the second vector representation through a second hybrid gating module of the recommendation model; and determining a recommended commodity corresponding to the user data according to the user vector representation and the commodity vector representation through a classification module of a recommendation model.

Description

technical field [0001] The present invention relates to the technical field of neural networks, and in particular, to a method and apparatus for determining a recommended commodity, an electronic device and a readable storage medium. Background technique [0002] In the prior art, in an e-commerce scenario, a recommender system selects commodities of interest to a user from a large number of candidate commodities. In turn, these products are recommended to users to promote the growth of business goals. Since the magnitude of candidate products usually reaches tens of millions or even hundreds of millions, a common practice in the industry is to divide the recommendation system into two stages: recall and sorting. The goal of the recall phase is to find a small subset of relevant candidate items, the candidate set, from a large item corpus. The goal of the sorting phase is to sort the candidate set. The quality of the candidate set will affect the ranking effect. Therefore...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06Q30/06G06F16/9535G06F16/958G06N20/00
CPCG06Q30/0631G06N20/00G06F16/9535G06F16/958
Inventor 胡奇夫薛岱月孙铭泽王伟鹏余建平
Owner BEIJING SANKUAI ONLINE TECH CO LTD