Text Geolocation Prediction via Self-Supervised Learning

Text geolocation prediction aims to infer the geographic location of text with text semantics, serving as a fundamental task for various geographic applications. As the mainstream approach, the deep learning-based methods follow the supervised learning paradigms, which rely heavily on a large amount...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuxing Wu, Zhuang Zeng, Kaiyue Liu, Zhouzheng Xu, Yaqin Ye, Shunping Zhou, Huangbao Yao, Shengwen Li
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:ISPRS International Journal of Geo-Information
Subjects:
Online Access:https://www.mdpi.com/2220-9964/14/4/170
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850144901443878912
author Yuxing Wu
Zhuang Zeng
Kaiyue Liu
Zhouzheng Xu
Yaqin Ye
Shunping Zhou
Huangbao Yao
Shengwen Li
author_facet Yuxing Wu
Zhuang Zeng
Kaiyue Liu
Zhouzheng Xu
Yaqin Ye
Shunping Zhou
Huangbao Yao
Shengwen Li
author_sort Yuxing Wu
collection DOAJ
description Text geolocation prediction aims to infer the geographic location of text with text semantics, serving as a fundamental task for various geographic applications. As the mainstream approach, the deep learning-based methods follow the supervised learning paradigms, which rely heavily on a large amount of labeled samples to train model parameters. To address this limitation, this paper presents a method for text geolocation prediction without labeled samples, namely GeoSG (Geographic Self-Supervised Geolocation) model, which leverages self-supervised learning to improve text geolocation prediction in situations where labeled samples are unavailable. Specifically, GeoSG integrates spatial distance and hierarchical constraints to characterize the interactions of POIs and text in a geographic relationship graph. And it designs two self-supervised tasks to train a shared network to learn the relationships among POIs and texts. Finally, the text geolocations are inferred based on the trained shared network. Experimental results on two datasets show that the proposed method outperforms the state-of-the-art baselines and is robust. This study provides a methodological reference for geolocating various text documents and offers a solution for numerous geographic intelligence tasks that lack labeled samples.
format Article
id doaj-art-b804da89f3724a0baf44c5a3f0c11f40
institution OA Journals
issn 2220-9964
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series ISPRS International Journal of Geo-Information
spelling doaj-art-b804da89f3724a0baf44c5a3f0c11f402025-08-20T02:28:14ZengMDPI AGISPRS International Journal of Geo-Information2220-99642025-04-0114417010.3390/ijgi14040170Text Geolocation Prediction via Self-Supervised LearningYuxing Wu0Zhuang Zeng1Kaiyue Liu2Zhouzheng Xu3Yaqin Ye4Shunping Zhou5Huangbao Yao6Shengwen Li7School of Computer Science, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430074, ChinaSchool of Computer Science, McGill University, Montréal, QC H3A 0E9, CanadaSchool of Computer Science, China University of Geosciences, Wuhan 430074, ChinaText geolocation prediction aims to infer the geographic location of text with text semantics, serving as a fundamental task for various geographic applications. As the mainstream approach, the deep learning-based methods follow the supervised learning paradigms, which rely heavily on a large amount of labeled samples to train model parameters. To address this limitation, this paper presents a method for text geolocation prediction without labeled samples, namely GeoSG (Geographic Self-Supervised Geolocation) model, which leverages self-supervised learning to improve text geolocation prediction in situations where labeled samples are unavailable. Specifically, GeoSG integrates spatial distance and hierarchical constraints to characterize the interactions of POIs and text in a geographic relationship graph. And it designs two self-supervised tasks to train a shared network to learn the relationships among POIs and texts. Finally, the text geolocations are inferred based on the trained shared network. Experimental results on two datasets show that the proposed method outperforms the state-of-the-art baselines and is robust. This study provides a methodological reference for geolocating various text documents and offers a solution for numerous geographic intelligence tasks that lack labeled samples.https://www.mdpi.com/2220-9964/14/4/170text geolocationself-supervised learninggraph neural networksdata-scarce scenarios
spellingShingle Yuxing Wu
Zhuang Zeng
Kaiyue Liu
Zhouzheng Xu
Yaqin Ye
Shunping Zhou
Huangbao Yao
Shengwen Li
Text Geolocation Prediction via Self-Supervised Learning
ISPRS International Journal of Geo-Information
text geolocation
self-supervised learning
graph neural networks
data-scarce scenarios
title Text Geolocation Prediction via Self-Supervised Learning
title_full Text Geolocation Prediction via Self-Supervised Learning
title_fullStr Text Geolocation Prediction via Self-Supervised Learning
title_full_unstemmed Text Geolocation Prediction via Self-Supervised Learning
title_short Text Geolocation Prediction via Self-Supervised Learning
title_sort text geolocation prediction via self supervised learning
topic text geolocation
self-supervised learning
graph neural networks
data-scarce scenarios
url https://www.mdpi.com/2220-9964/14/4/170
work_keys_str_mv AT yuxingwu textgeolocationpredictionviaselfsupervisedlearning
AT zhuangzeng textgeolocationpredictionviaselfsupervisedlearning
AT kaiyueliu textgeolocationpredictionviaselfsupervisedlearning
AT zhouzhengxu textgeolocationpredictionviaselfsupervisedlearning
AT yaqinye textgeolocationpredictionviaselfsupervisedlearning
AT shunpingzhou textgeolocationpredictionviaselfsupervisedlearning
AT huangbaoyao textgeolocationpredictionviaselfsupervisedlearning
AT shengwenli textgeolocationpredictionviaselfsupervisedlearning