A POI recommendation method and device based on spatio-temporal sequence and social embedding ranking

A point of interest and sequence technology, applied in the field of deep learning, can solve problems such as poor recommendation performance, and achieve the effect of accurate recommendation

Active Publication Date: 2022-03-15
WUHAN UNIV
View PDF25 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In view of this, the present invention provides a point-of-interest recommendation method and device based on spatio-temporal sequence and social embedding ranking to solve or at least partially solve the technical problem of poor recommendation performance existing in the methods in the prior art

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
  • A POI recommendation method and device based on spatio-temporal sequence and social embedding ranking
  • A POI recommendation method and device based on spatio-temporal sequence and social embedding ranking
  • A POI recommendation method and device based on spatio-temporal sequence and social embedding ranking

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0055] This embodiment provides a point-of-interest recommendation method based on spatio-temporal sequence and social embedding ranking, the method includes:

[0056] Step S1: Use the hybrid deep model based on horizontal convolution filter, vertical convolution filter and multi-layer perceptron model to model the user check-in timing information, learn the objective function of the timing features of the points of interest, and obtain the user check-in timing model.

[0057] Specifically, based on the hybrid deep model of horizontal convolution filter, vertical convolution filter and multi-layer perceptron model, the spatio-temporal sequence and check-in information of the user's check-in position are first embedded into the latent dimension space, and the embedded information The matrix is ​​regarded as a series of "images", and the time-space sequence mode and user check-in information are regarded as the local features of the "image", and then the joint convolution filter ...

Embodiment 2

[0171] Based on the same inventive concept as in Embodiment 1, this embodiment also provides a device for recommending points of interest based on time-space sequence and social embedding ranking, please refer to Figure 4 , the device consists of:

[0172] The user check-in timing model building module 201 is used to model the user check-in timing information using a hybrid depth model based on horizontal convolution filters, vertical convolution filters and multi-layer perceptron models, and learn the objective function of the timing features of points of interest , get the user sign-in timing model;

[0173] The user social information model construction model 202 is used to construct a weight function based on the metric learning theory for predicting the degree of social relationship between users and users, model the user social information, and obtain the user social information model;

[0174] The information fusion module 203 is used to fuse the user check-in timing ...

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 point-of-interest recommendation method based on spatio-temporal sequence and social embedding ranking. First, a hybrid depth model based on horizontal convolution filter, vertical convolution filter and multi-layer perceptron model is proposed to capture User preferences and the impact of spatio-temporal sequence patterns; then, the metric learning method is used to model the user's social relationship; then, a unified framework based on matrix decomposition is used to integrate the user's personal interests, check-in sequence patterns, and user social information. Finally, the BPR standard is used to optimize the target loss function to fit the partial order relationship of the user on the POI pair, and finally generate the POI recommendation list. The present invention makes the final recommendation more accurate by constructing a point-of-interest recommendation method that combines a hybrid deep model and metric learning.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a point-of-interest recommendation method and device based on spatio-temporal sequence and social embedding ranking. Background technique [0002] With the rapid development of Web2.0, wireless communication and location acquisition technology have promoted many location-based social network applications. In these location-based social network service applications, users can establish social connections with other users and explore the surrounding environment. Share their life experiences and experiences by checking in points of interest (such as restaurants, shopping malls and attractions, etc.). In addition to providing an interactive platform for users, LBSN contains rich data (check-in data, social relations, comment information, etc.) that can be used to mine users' interests and preferences, and recommend unvisited geographic locations that users may be intere...

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 Patents(China)
IPC IPC(8): G06F16/9535G06N3/04G06N3/08
CPCG06F16/9535G06N3/08G06N3/048G06N3/045
Inventor 李雪飞徐洋洋高榕张玉洁饶建勋
Owner WUHAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap