408 Hits in 6.9 sec

A Deep Learning Approach to Geographical Candidate Selection through Toponym Matching [article]

Mariona Coll Ardanuy, Kasra Hosseini, Katherine McDonough, Amrey Krause, Daniel van Strien, Federico Nanni
2020 arXiv   pre-print
In this paper, we introduce a flexible deep learning method for candidate selection through toponym matching, using state-of-the-art neural network architectures.  ...  Candidate selection is the task of identifying the potential entities that can be referred to by a toponym previously recognized.  ...  In this paper, we present a new and flexible deep learning approach to geographical candidate selection through toponym matching, which is specifically tailored to dealing with these challenges characteristic  ... 
arXiv:2009.08114v2 fatcat:2eczu5x4rvbmfjbfttfrnglzpi

How can voting mechanisms improve the robustness and generalizability of toponym disambiguation? [article]

Xuke Hu, Yeran Sun, Jens Kersten, Zhiyong Zhou, Friederike Klan, Hongchao Fan
2022 arXiv   pre-print
Recently, many novel approaches have been proposed, especially deep learning-based approaches, such as CamCoder, GENRE, and BLINK.  ...  Extracting geographic information from texts is called Geoparsing, which includes two subtasks: toponym recognition and toponym disambiguation, i.e., to identify the geospatial representations of toponyms  ...  Rules Given a target toponym, rule-based approaches first search gazetteers to find all the candidate entries that match or partially match the string of the toponym, and then rank or score the candidate  ... 
arXiv:2209.08286v1 fatcat:ui3swgviyfgrlj63jpxzwq3ihi

Geocoding location expressions in Twitter messages: A preference learning method

Judith Gelernter, Wei Zhang
2014 Journal of Spatial Information Science  
We used supervised machine learning to weigh the different fields of the Twitter message and the features of a world gazetteer to create a model that will prefer the correct gazetteer candidate to resolve  ...  Correct resolution is made even more difficult when there is little context to determine which place is intended, as in a 140-character Twitter message, or when location cues from different sources conflict  ...  Acknowledgments We thank Brendan O'Conner for access to the stored tweets from his archive, downloaded from the Twitter Gardenhose/Decahose, at Carnegie Mellon University.  ... 
doi:10.5311/josis.2014.9.170 fatcat:yd7npoy23jgypnzc5mvgy3eh6u

A Novel Deep Learning Approach Using Contextual Embeddings for Toponym Resolution

Ana Bárbara Cardoso, Bruno Martins, Jacinto Estima
2021 ISPRS International Journal of Geo-Information  
This article describes a novel approach for toponym resolution with deep neural networks.  ...  The proposed approach does not involve matching references in the text against entries in a gazetteer, instead directly predicting geo-spatial coordinates.  ...  The funders also had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the or in the decision to publish the results.  ... 
doi:10.3390/ijgi11010028 fatcat:g7nfjp33bffvhdi5wp24qrefya

Deep Learning for Toponym Resolution: Geocoding Based on Pairs of Toponyms

Jacques Fize, Ludovic Moncla, Bruno Martins
2021 ISPRS International Journal of Geo-Information  
The proposed approach is based on a deep neural network that uses Long Short-Term Memory (LSTM) units to produce representations from sequences of character n-grams.  ...  In this work, we propose a geocoding approach based on modeling pairs of toponyms, which returns latitude-longitude coordinates.  ...  This data can be found here: (accessed on 28 November 2021). Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/ijgi10120818 fatcat:sttddeumfbg4lnkjm4tptfrzr4

An Automated Approach for Geocoding Tabular Itineraries

Rui Santos, Patricia Murrieta-Flores, Bruno Martins
2017 Proceedings of the 11th Workshop on Geographic Information Retrieval - GIR'17  
The obtained results show that while approximate string matching can already achieve very low median errors, with many toponyms matching exactly against GeoNames entries, the combination with cost optimization  ...  This article advances a novel method for automatically geocoding tabular itineraries, combining approximate string matching with a cost optimization algorithm based on dynamic programming.  ...  The researchers from INESC-ID also had financial dupport from Fundação para a Ciência e Tecnologia (FCT), through the project grant with reference PTDC/EEI-SCR/1743/2014 (Saturn), as well as through the  ... 
doi:10.1145/3155902.3155908 dblp:conf/gir/SantosMM17 fatcat:jmmrx5n35bfb3mag4zohjvf7ca

Location reference recognition from texts: A survey and comparison [article]

Xuke Hu, Zhiyong Zhou, Hao Li, Yingjie Hu, Fuqiang Gu, Jens Kersten, Hongchao Fan, Friederike Klan
2022 arXiv   pre-print
, statistical learning-based, and hybrid approaches.  ...  To fill these research gaps, this review first summarizes seven typical application domains of geoparsing: geographic information retrieval, disaster management, disease surveillance, traffic management  ...  Magge et al. [117] used a deep feedforward neural network to determine whether a given phrase in biomedical articles is a toponym or not.  ... 
arXiv:2207.01683v1 fatcat:xiy7az4veza6lm52c6yj7mnoe4

Geoparsing: Solved or Biased? An Evaluation of Geographic Biases in Geoparsing

Zilong Liu, Krzysztof Janowicz, Ling Cai, Rui Zhu, Gengchen Mai, Meilin Shi
2022 AGILE: GIScience Series  
Geoparsing, the task of extracting toponyms from texts and associating them with geographic locations, has witnessed remarkable progress over the past years.  ...  Particularly, our spatially-explicit performance evaluation serves as an approach to evaluation bias mitigation in geoparsing.We conclude that previous performance evaluations published in the literature  ...  Each recognized toponym is then fed into a toponym resolution model that selects the correct place reference along with its geo-location information from all candidates with the same place name.  ... 
doi:10.5194/agile-giss-3-9-2022 fatcat:4h5wuwy32zdvtlrb75fz4qeoky

Toponym Extraction and Disambiguation Enhancement Using Loops of Feedback [chapter]

Mena B. Habib, Maurice van Keulen
2013 Communications in Computer and Information Science  
We conducted experiments with a set of holiday home descriptions with the aim to extract and disambiguate toponyms.  ...  In this paper we aim to investigate both avenues and to show that explicit handling of the uncertainty of annotation has much potential for making both extraction and disambiguation more robust.  ...  Toponym Disambiguation For the toponym disambiguation task, we only select those toponyms annotated by the extraction models that match a reference in GeoNames.  ... 
doi:10.1007/978-3-642-54105-6_8 fatcat:p52ig6oqybhrtnniafr647yekm

Machine learning for cross-gazetteer matching of natural features

Elise Acheson, Michele Volpi, Ross S. Purves
2019 International Journal of Geographical Information Science  
We argue that future work in this area should strive to be more reproducible and report results on a realistic testing pipeline including candidate selection, feature extraction, and classification.  ...  We first perform rule-based matching, establishing competitive results, then apply machine learning using Random Forests, a method well-suited to the matching task.  ...  His main research activities are at the interface of computer vision, machine learning, and deep learning, focusing on the extraction of information from aerial and satellite imagery and from geospatial  ... 
doi:10.1080/13658816.2019.1599123 fatcat:qr5p2hhbmzdf5n4uhn6gfd326i

Extracting locations from sport and exercise-related social media messages using a neural network-based bilingual toponym recognition model

Pengyuan Liu, Sonja Koivisto, Tuomo Hiippala, Charlotte Van der Lijn, Tuomas Vaisanen, Marisofia Nurmi, Tuuli Toivonen, Kirsi Vehkakoski, Janne Pyykonen, Ilkka Virmasalo, Mikko Simula, Elina Hasanen (+2 others)
2022 Journal of Spatial Information Science  
We show that our approach outperforms five state-of-the-art deep learning and machine learning models.  ...  To support this effort, this article presents an end-to-end deep learning-based bilingual toponym recognition model for extracting location information from social media content related to sports and exercise  ...  Acknowledgments This study is a part of the "Equality in suburban physical activity environments, YLLI" research project (in Finnish: Yhdenvertainen liikunnallinen lähiö, YLLI).  ... 
doi:10.5311/josis.2022.24.167 fatcat:e57deolspfedlbwe7gmnffxabi

Event Geoparser with Pseudo-Location Entity Identification and Numerical Argument Extraction Implementation and Evaluation in Indonesian News Domain

Agung Dewandaru, Dwi Hendratmo Widyantoro, Saiful Akbar
2020 ISPRS International Journal of Geo-Information  
Geoparser is a fundamental component of a Geographic Information Retrieval (GIR) geoparser, which performs toponym recognition, disambiguation, and geographic coordinate resolution from unstructured text  ...  As a side effect of event extraction, various numerical arguments are also extracted, and the output is easily projected to a rich choropleth map from a single news document.  ...  Machine learning approach to detect event triggers has been done, for example, by [55] .  ... 
doi:10.3390/ijgi9120712 fatcat:qvd2ltdwtjabdjwrh47oxnqrhe

What's missing in geographical parsing?

Milan Gritta, Mohammad Taher Pilehvar, Nut Limsopatham, Nigel Collier
2017 Language Resources and Evaluation  
The ability to geo-locate events in textual reports represents a valuable source of information in many real-world applications such as emergency responses, real-time social media geographical event analysis  ...  Geographical data can be obtained by converting place names from free-format text into geographical coordinates.  ...  Given a list of candidate coordinates for each location and the surrounding context, the goal is to select the correct coordinate i.e. disambiguate.  ... 
doi:10.1007/s10579-017-9385-8 pmid:31258456 pmcid:PMC6560650 fatcat:ikvd6b5lgrhpjkvdjfrtbil7ya

Towards geo-referencing infrastructure for local news

Guoray Cai, Ye Tian
2016 Proceedings of the 10th Workshop on Geographic Information Retrieval - GIR '16  
LocusRecommender automatically suggests the best matches from gazetteer ranked by a set of heuristic rules.  ...  To gain insights on the unique aspects of local gazetteers and the nature of ambiguities, we present an analysis of a collection of local new articles.  ...  Acknowledgment This work is partially supported by a grant from the National Science Foundation under award IIS-1211059.  ... 
doi:10.1145/3003464.3003473 dblp:conf/gir/CaiT16 fatcat:fbpowyyun5abjglmd4d5du7yhe

Automatic Identification of Addresses: A Systematic Literature Review

Paula Cruz, Leonardo Vanneschi, Marco Painho, Paulo Rita
2021 ISPRS International Journal of Geo-Information  
The main findings revealed a consistent move from more traditional approaches to deep learning methods based on semantics, encoder-decoder architectures, and attention mechanisms, as well as the very recent  ...  Address matching continues to play a central role at various levels, through geocoding and data integration from different sources, with a view to promote activities such as urban planning, location-based  ...  The deep neural network based on GRUs, to categorize toponym pairs as matches or non-matches, proposed by Santos et al.  ... 
doi:10.3390/ijgi11010011 fatcat:z56kyinqfnbbjbv63jnkxtsg4i
« Previous Showing results 1 — 15 out of 408 results