Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Commodity recommendation method and system based on user session and graph convolutional neural network

A convolutional neural network and product recommendation technology, applied in the field of user preference product recommendation, can solve problems such as long time, many training parameters, and reduced product recommendation efficiency, and achieve the effect of improving accuracy and speed, and improving accuracy

Active Publication Date: 2019-11-22
QILU UNIV OF TECH
View PDF3 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the technical level, the inventors found that using a recurrent neural network would be limited by the number of user behaviors within the session, and when the user's click records were too few or too many, the effect of the recommendation model would be affected; while using Markov When using a model, it only models the one-way transfer relationship between two adjacent commodities, while ignoring other commodities in the session; and the neural network shows problems such as many training parameters and long time in model training, and this To a certain extent, it limits the application of conversational recommendation in neural network and reduces the efficiency of product recommendation

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
  • Commodity recommendation method and system based on user session and graph convolutional neural network
  • Commodity recommendation method and system based on user session and graph convolutional neural network
  • Commodity recommendation method and system based on user session and graph convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] figure 1 The flow chart of the product recommendation method based on user session and graph convolutional neural network in this embodiment is given.

[0030] to combine figure 1 , the product recommendation method based on user sessions and graph convolutional neural networks in this embodiment includes:

[0031] S101: Receive the clicked commodity sequence in the same time unit, model it as a session in the form of a directed graph, and construct a session graph; wherein, the session graph is expressed in the form of an embedding vector.

[0032] In step S101, the specific process of constructing the session graph is as follows:

[0033] Obtain a click sequence of the user within the same time unit, forming a session. For example, if the click sequence generated by a user on a shopping website on July 21 is (dress→high heels→necklace→skirt), then such a click sequence becomes a session.

[0034] Use the divided session to model in the form of a directed graph, su...

Embodiment 2

[0114] figure 2 A schematic structural diagram of the product recommendation system based on user sessions and graph convolutional neural networks in this embodiment is given.

[0115] combine figure 2 , the product recommendation system based on user sessions and graph convolutional neural networks in this embodiment includes:

[0116] (1) Conversation graph building module, which is used to receive the clicked product sequence in the same time unit, as a session and modeled in the form of a directed graph, to construct a session graph; wherein, the session graph is expressed in the form of an embedded vector ;

[0117] Specifically, in the session graph construction module, each node in the session graph represents a commodity, each edge is the order in which the commodities are clicked, and the weight of each edge is equal to the number of occurrences of the edge divided by the starting point of the edge. The out-degree of the starting node; each product is embedded in...

Embodiment 3

[0141] This embodiment provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps in the product recommendation method based on user session and graph convolutional neural network as described in Embodiment 1 are realized. .

[0142]In this embodiment, by constructing a session graph, comprehensively paying attention to the user's long-term and short-term preferences and fluctuating discrete preferences, considering the internal order of each session of the user and the correlation between multiple sessions, it is possible to consider complex data forms and Network structure, input the conversation graph into the gated graph neural network, output the posterior probabilities of all commodities contained in the conversation graph, use the magnitude of the posterior probability to output commodity recommendation results, and improve the The accuracy and speed of product recommendation are improved...

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 provides a commodity recommendation method and system based on a user session and a graph convolutional neural network. The commodity recommendation method comprises the following steps:receiving a clicked commodity sequence in the same time unit, taking the clicked commodity sequence as a session, modeling in the form of a directed graph, and constructing a session graph, wherein the session graph is expressed in the form of an embedded vector; inputting the plurality of session graphs in the embedded vector form into a gated graph neural network, and outputting posterior probabilities of all commodities contained in the session graphs; and according to the descending order, screening out the commodities corresponding to the posterior probabilities ranked at the front preset digits as user preference commodity prediction results, and recommending the commodities one by one. On the premise of considering the commodity relevance, the commodity recommendation accuracy andspeed are improved.

Description

technical field [0001] The present disclosure belongs to the field of product recommendation for user preference, and in particular relates to a method and system for product recommendation based on user session and graph convolutional neural network. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] Most of the current recommendation methods for processing dialogue use recurrent neural networks and Markov chains. At the technical level, the inventors found that using a recurrent neural network would be limited by the number of user behaviors within the session, and when the user's click records were too few or too many, the effect of the recommendation model would be affected; while using Markov When using a model, it only models the one-way transfer relationship between two adjacent commodities, while ignoring other commodities in the se...

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
IPC IPC(8): G06Q30/06G06F16/9535G06F16/9537G06F16/36G06N3/04
CPCG06Q30/0631G06F16/9535G06F16/9537G06F16/367G06N3/045
Inventor 杨振宇张鸣鸽
Owner QILU UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products