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

Point-of-interest combination recommendation algorithm based on location social network

A social network and point of interest technology, applied in the field of big data recommendation models, can solve the problems of not discussing the impact of geographical correlation on user preferences, lack of geographical attributes, and inability to fully express user preference information

Pending Publication Date: 2022-07-08
NORTHWESTERN POLYTECHNICAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing combination model lacks in-depth research on geographic attributes, and does not explore the impact of geographic correlation on user preferences, and the mobile trajectory only exposes part of the observation results of the entire trajectory, which cannot fully represent the user's preference information.

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
  • Point-of-interest combination recommendation algorithm based on location social network
  • Point-of-interest combination recommendation algorithm based on location social network
  • Point-of-interest combination recommendation algorithm based on location social network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] 1. Calculate the user's collaborative preference score for POIs.

[0047] Based on the open source Gowalla dataset and Foursquare dataset, first, the interaction graph of user interest points is used as the input of the lightweight graph convolutional network LightGCN, e u (0) represents the ID embedding of user u, e p (0) Representing the ID embedding of the interest point p, the propagation rules of the embedding representation in LightGCN are defined as follows:

[0048]

[0049]

[0050] N u represents the set of interest points that user u interacts with, N p represents the set of users interacting with the point of interest p, is the symmetric normalization regularization term, which can avoid the increase of the embedding size with the increase of graph convolution operations.

[0051] In LightGCN, the only trainable model parameter is the embedding of layer 0, i.e. e u (0) and e i (0) . The higher-level embeddings can be calculated through the pr...

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 an interest point combination recommendation algorithm based on a location social network. The method comprises the following steps: (1) obtaining collaborative preference scores of a user on interest points through a LightGCN graph convolutional network model; (2) updating the embedded representation of the interest points by using the geographic correlation between the interest points; (3) obtaining geographical preference scores of the user on the interest points through a GRU sequence model with an attention mechanism; and (4) obtaining a final interest point recommendation score through a linear weighted combination strategy. According to the method, geographic information, cooperation information and sequence information are comprehensively considered, and the recall rate of offline recommendation is increased.

Description

technical field [0001] The invention relates to the field of data mining, in particular to a big data recommendation model. Background technique [0002] With the rapid development of the Internet, personalized recommendation services are becoming more and more important. More and more people use the Internet with the help of smart devices such as mobile phones. On the one hand, it realizes the intelligentization of life scenes, and on the other hand, it inevitably brings about the problem of information explosion. Faced with massive amounts of data, it is a challenging task for users to quickly and accurately retrieve the information they are interested in. The recommender system can establish a user interest model by analyzing the user's historical behavior under the premise that the user does not clearly indicate their own needs, and recommends points of interest that meet the user's interest. In the face of various application scenarios, a single recommendation method ...

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): G06F16/9535G06F16/9537G06N3/04G06N3/08G06Q50/00
CPCG06F16/9535G06F16/9537G06N3/08G06Q50/00G06N3/048G06N3/044G06N3/045
Inventor 慕志颖徐莎莎李晓宇
Owner NORTHWESTERN POLYTECHNICAL UNIV
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