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

Recommendation method and system based on tree structure, medium and electronic equipment

A recommendation method and recommendation system technology, applied in the field of recommendation based on tree structure, can solve the problems that the TDM model is difficult to accurately describe the relationship between products and products, and it is difficult for users to recommend more suitable products, so as to improve expression ability and train efficient Effect

Pending Publication Date: 2020-04-28
CTRIP COMP TECH SHANGHAI
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0011] The technical problem to be solved by the present invention is to overcome the defect that the TDM model in the prior art is difficult to accurately describe the relationship between commodities, making it difficult to recommend more suitable products to users, and to provide a tree-based recommendation method, Systems, media and electronics

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
  • Recommendation method and system based on tree structure, medium and electronic equipment
  • Recommendation method and system based on tree structure, medium and electronic equipment
  • Recommendation method and system based on tree structure, medium and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] like figure 1 As shown, a tree-based recommendation method includes the following steps:

[0049] S1. Generate a sequence of files from the historical data of the user's access to the commodity;

[0050] The historical data includes one or more items of the user's clicked product records, favorited product records, ordered product records, and product information. This historical data will be used as the data for item pre-training and the source data for neural network model training.

[0051] According to the time when the user accesses the product, the historical records of the user's clicks, favorites, and orders, such as product numbers and hotel numbers, are organized into a comma-separated sequence in order for the input of the glove model. The corresponding commodity information can also be used as part of the training data. Compared with the input data volume of the original TDM model, this historical record adds attributes such as commodity prices other than ...

Embodiment 2

[0075] A tree-based recommendation system, such as figure 2 shown, including:

[0076] Sequence file generation module 1, used to generate a sequence file from the historical data of the user's access to commodities;

[0077] The first training module 2 is used to train the sequence file through the glove model, so as to obtain the embedding vector of the commodity;

[0078] A tree model building module 3, configured to use the embedding vector to build a tree model of the commodity;

[0079] The tree model includes a parent node and a leaf node, and the embedding vector of the parent node is the mean or weighted mean of the leaf nodes.

[0080] A data set generation module 4, configured to generate a data set from the historical data according to the tree model;

[0081] In the data set, the commodities with positive feedback from each user are taken as positive samples, and the commodities that other users have not interacted with are sampled to generate negative samples...

Embodiment 3

[0088] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the tree-structure-based recommendation method provided in Embodiment 1 are implemented.

[0089] Wherein, the readable storage medium may more specifically include but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device or any of the above-mentioned the right combination.

[0090] In a possible implementation manner, the present invention can also be implemented in the form of a program product, which includes program code, and when the program product runs on the terminal device, the program code is used to make the terminal device execute Steps of the tree structure-based recommendation method in Embodiment 1.

[0091] Wherein, the program code for executing the present invention can be written i...

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 recommendation method and system based on a tree structure, a medium and electronic equipment. The method comprises the following steps: generating a sequence file from historical data of commodities accessed by a user; training the sequence file through a glove model to obtain an embedded vector of the commodity; constructing a tree model of the commodity according to the embedded vector; generating a data set from the historical data according to the tree model; training a neural network model by using the data set; and recalling the tree model layer by layer through the trained neural network model to recommend commodities. According to the scheme, the relationship between the commodities can be accurately described, more suitable products can be recommended tothe user, and global article recommendation recall can be efficiently carried out on the premise of a high recall rate.

Description

technical field [0001] The invention relates to a tree structure-based recommendation method. Background technique [0002] With the continuous development of Internet-related services, the complexity of Internet services continues to increase. Whether it is an e-commerce platform that mainly sells commodities, an OTA (Online Travel Agency) website that provides travel products and services, or a website that provides various content information, the overall number of commodities (items) it can provide is far from Exceeded the maximum number that the user can handle. [0003] In order to deal with such problems, various recommendation algorithms have emerged. The biggest function of the recommendation algorithm is that it helps users to filter the items they are interested in first, reducing the time and cost for users to retrieve the required information. In addition, the recommendation system has also become one of the most important infrastructures of Internet service ...

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): G06F16/9535G06F16/901G06N3/04G06N3/08
CPCG06F16/9535G06F16/9027G06N3/082G06N3/045
Inventor 宣云儿
Owner CTRIP COMP TECH SHANGHAI
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com