DOCSLIB.ORG

Toponym Resolution in Text Information Systems and Computer

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Toponym Resolution in Text Information Systems and Computer Toponym Resolution in Text Ana Bárbara Inácio Cardoso Thesis to obtain the Master of Science Degree in Information Systems and Computer Engineering Supervisors: Prof. Bruno Emanuel da Graça Martins Prof. Jacinto Paulo Simões Estima Examination Committee Chairperson: Prof. Alexandre Paulo Lourenço Francisco Supervisor: Prof. Bruno Emanuel da Graça Martins Members of the Committee: Prof. Fernando Manuel Marques Batista October 2019 Acknowledgements First of all, I would like to thank my advisors Professor Bruno Emanuel da Gra¸caMartins and Professor Jacinto Paulo Sim~oesEstima, for their guidance during the development of this work and their immense contributions to the growth and success of this thesis. I would like to thank my amazing family for their support, motivation, and for allowing me to learn and to grow professionally and personally during the time spent in Instituto Superior T´ecnico.Thank you for being so supportive. I am also grateful for the support and constant motivation of my friends and colleagues during this entire journey over the past five years. Thank you for the fantastic time, joys, victories, and all the tears that were shared. Finally, I would like to thank the Funda¸c~aopara a Ci^enciae Tecnologia (FCT), for sup- porting the work developed during this dissertation, through the project grants with ref- erences PTDC/EEI-SCR/1743/2014 (Saturn), T-AP HJ-253525 (DigCH), and PTDC/CCI- CIF/32607/2017 (MIMU), as well as through the INESC-ID multi-annual funding from the PIDDAC programme (UID/CEC/5 0021/2019). I also gratefully acknowledge the support of NVIDIA Corporation, with the donation of two Titan Xp GPUs used in the reported experi- ments. Ana B´arbaraIn´acio Cardoso For my family, Resumo A resolu¸c~aode top´onimosem texto, em que um top´onimose refere a um nome de local ou a uma refer^enciade local, consiste na desambigua¸c~aodestas refer^encias,associando a cada uma delas uma localiza¸c~ao´unica,portanto inequ´ıvoca, sobre a superf´ıcieda Terra (por exem- plo, atrav´esda atribui¸c~aode coordenadas geogr´aficasde latitude e longitude). Dado que os nomes dos locais s~aoaltamente amb´ıguos,a resolu¸c~aode top´onimos´euma tarefa desafiante; por exemplo, existem m´ultiploslocais na Terra que partilham o mesmo nome, e ainda m´ultiplas des- igna¸c~oesque referem o mesmo local. A resolu¸c~aode top´onimos´euma tarefa muito interessante, uma vez que v´ariasaplica¸c~oesposs´ıveis podem beneficiar dos resultados, incluindo o apoio ao processamento e an´alisede informa¸c~aogeogr´aficapresente em cole¸c~oesextensas de documentos, assim como o suporte `ageolocaliza¸c~aode documentos. O trabalho que desenvolvi durante a tese de mestrado, descrito neste documento de disserta¸c~ao,teve como objetivo a an´alisede estudos desenvolvidos na ´areaat´eao momento, bem como o desenvolvimento de um modelo para a resolu¸c~aode top´onimosconsiderando t´ecnicasdo estado-da-arte aplicadas ao processamento de l´ınguanatural. A arquitetura de rede neural proposta utiliza unidades recorrentes com m´ultiplas entradas (por exemplo, o top´onimoa ser desambiguado juntamente com as palavras adjacentes), aproveitando especificamente incorpora¸c~oesde palavras contextuais pr´e-treinadas(incorpora¸c~oes ELMo ou BERT) e unidades bidirecionais de Long Short-Term Memory (LSTM), ambas muito utilizadas para a modela¸c~aode dados textuais. Adicionalmente, o modelo proposto foi avali- ado em diferentes contextos, (i) usando informa¸c~oesexternas extra´ıdasde dados rasterizados com informa¸c~oesgeof´ısicas,incluindo cobertura terrestre, eleva¸c~aodo terreno, entre outras, e (ii) usando dados adicionais de artigos da Wikip´ediaem ingl^espara treinar o modelo com o objetivo de guiar e ajudar durante o treino. Os resultados obtidos mostram uma qualidade significativamente superior do m´etodo proposto, em compara¸c~aocom as abordagens anteriores, particularmente no cen´arioque envolve incorpora¸c~oesBERT juntamente com a adi¸c~aode dados. Abstract Toponym resolution in text, where toponym refers to a place name or place reference, consists in the disambiguation of these references, by associating each one of them to a unique, thus unambiguous, location over the surface of the Earth (e.g., through the assignment of latitude and longitude geographical coordinates). Given that place names are highly ambiguous, the toponym resolution is a challenging task; for instance, multiple places on Earth share the same name and even multiple designations that refer to the same location. Toponym resolution is an exciting task since several possible applications can benefit from the results, which includes the support of the processing and analysis of geographic information present in collections of large documents, as well as the support of document geolocation. The research conducted during the MSc thesis, and presented in this dissertation aims to analyze the studies developed in the area so far, as well as the development of a model for the toponym resolution considering state-of-the-art techniques applied to natural language processing. The proposed neural network architecture uses recurrent units with multiple inputs (e.g., the toponym to disambiguate along with the surrounding words), leveraging pre-trained contextual word embeddings (i.e., ELMo or BERT embeddings) and bi-directional Long Short-Term Memory (LSTM) units, both commonly used for textual data modeling. Additionally, the proposed model was evaluated in different contexts, (i) using external information extracted from raster data with geophysical information, including land cover, terrain elevation, among others, and (ii) using additional data from English Wikipedia articles to train the model, to guide and help during the model training. The obtained results show a significantly higher quality of the proposed method, in comparison to previous approaches and particularly in the setting that involves BERT embeddings and additional data. Palavras-chave Keywords Palavras-chave An´alisegeogr´aficade texto Resolu¸c~aode top´onimosem texto Aprendizagem profunda aplicada ao Processamento de L´ınguaNatural Redes neuronais recorrentes Representa¸c~oescontextuais de incorpora¸c~aode palavras Propriedades geof´ısicas Keywords Geographical text analysis Toponym resolution in text Deep learning for Natural Language Processing Recurrent neural networks Contextual word embedding representations Geophysical properties Contents 1 Introduction 1 1.1 Motivation . .1 1.2 Thesis Proposal . .2 1.3 Contributions . .3 1.4 Structure of the Document . .4 2 Fundamental Concepts 5 2.1 Introduction to Neural Networks and Deep Learning . .5 2.1.1 Feed-forward Neural Networks . .6 2.1.2 Optimization Algorithms to Neural Networks . .9 2.2 Recurrent Neural Networks . 11 2.2.1 Simple Recurrent Neural Network Architecture . 13 2.2.2 Long Short-Term Memory Architecture . 14 2.2.3 Gated Recurrent Unit Architecture . 15 2.3 Text Representation Methods . 15 2.3.1 Traditional Approaches . 16 2.3.2 Word Embeddings . 16 2.3.3 Contextual Word Embeddings . 18 2.3.3.1 Embeddings from Language Models . 19 2.3.3.2 Bidirectional Encoder Representations from Transformers . 20 2.4 Overview . 21 i 3 Related Work 23 3.1 Named Entity Linking . 23 3.2 Fine-Grained Entity Classification . 26 3.3 Toponym Resolution . 28 3.3.1 Heuristic Methods . 28 3.3.2 Combining Heuristics through Supervised Learning . 29 3.3.3 Methods Combining Geodesic Grids and Language Models . 30 3.3.4 Deep Learning Techniques . 30 3.4 Overview . 31 4 Toponym Resolution in Text 33 4.1 Toponym Resolution as Classification . 33 4.2 Proposed Model Architecture . 34 4.3 Additional Experiments with the Proposed Model . 37 4.3.1 Wikipedia instances . 38 4.3.2 Geophysical properties . 39 4.4 Overview . 40 5 Experimental Evaluation 41 5.1 Corpora Used in the Experiments . 41 5.2 Experimental Methodology . 42 5.3 The Obtained Results . 44 5.4 Overview . 46 6 Conclusions and Future Work 49 6.1 Overview on the Contributions . 50 6.2 Future Work . 51 Bibliography 57 ii List of Figures 2.1 Perceptron architecture. .6 2.2 Feed-forward neural network architecture. .7 2.3 Recurrent neural network architecture. 11 2.4 Multi-layer RNN architecture . 12 2.5 Bi-directional recurrent neural network. 13 2.6 Traditional text representation. 16 2.7 ELMo model. 19 2.8 BERT model. 20 4.1 HEALPix partitioning. 34 4.2 Proposed neural network architecture. 35 4.3 Using Wikipedia to create new data instances. 38 iii iv List of Tables 3.1 Methods used in previous toponym resolution systems. 32 4.1 Experiences with Wikipedia instances. 37 4.2 Land coverage classes. 39 5.1 Statistical characterization of the corpora. 42 5.2 Experiments with the proposed model. 43 5.3 Additional experiments results. 44 5.4 Locations and respective prediction distance error. 46 5.5 Illustrative examples. 47 v vi Introduction1 Toponymy resolution concerns the disambiguation of place names and other references to places in textual documents. The disambiguation is achieved by associating each of these place references to a unique position on the Earth's surface, e.g., through the assignment of geographic coordinates. Through the results emerging from toponymy resolution, it is possible to consider several applications, such as the improvement of search engine results (e.g., by geographic indexing or classification), document classification according to spatial criteria, which allows grouping documents into meaningful clusters and enables the mapping of textually encoded informa- tion (Monteiro et al., 2016). Another possible application is in areas such as computational social sciences or digital humanities (Wing, 2015), for instance, through supporting the auto- matic processing and analysis of geographic data encoded over extensive collections of textual documents. Moreover, place reference resolution can be an auxiliary component for the complete geolocation of documents (Melo and Martins, 2017), whereas the toponyms mentioned in the text can provide indications about the overall location of the document.
Load more

Recommended publications
  • Adaptive Geoparsing Method for Toponym Recognition and Resolution in Unstructured Text
    remote sensing Article Adaptive Geoparsing Method for Toponym Recognition and Resolution in Unstructured Text Edwin Aldana-Bobadilla 1,∗ , Alejandro Molina-Villegas 2 , Ivan Lopez-Arevalo 3 , Shanel Reyes-Palacios 3, Victor Muñiz-Sanchez 4 and Jean Arreola-Trapala 4 1 Conacyt-Centro de Investigación y de Estudios Avanzados del I.P.N. (Cinvestav), Victoria 87130, Mexico; [email protected] 2 Conacyt-Centro de Investigación en Ciencias de Información Geoespacial (Centrogeo), Mérida 97302, Mexico; [email protected] 3 Centro de Investigación y de Estudios Avanzados del I.P.N. Unidad Tamaulipas (Cinvestav Tamaulipas), Victoria 87130, Mexico; [email protected] (I.L.); [email protected] (S.R.) 4 Centro de Investigación en Matemáticas (Cimat), Monterrey 66628, Mexico; [email protected] (V.M.); [email protected] (J.A.) * Correspondence: [email protected] Received: 10 July 2020; Accepted: 13 September 2020; Published: 17 September 2020 Abstract: The automatic extraction of geospatial information is an important aspect of data mining. Computer systems capable of discovering geographic information from natural language involve a complex process called geoparsing, which includes two important tasks: geographic entity recognition and toponym resolution. The first task could be approached through a machine learning approach, in which case a model is trained to recognize a sequence of characters (words) corresponding to geographic entities. The second task consists of assigning such entities to their most likely coordinates. Frequently, the latter process involves solving referential ambiguities. In this paper, we propose an extensible geoparsing approach including geographic entity recognition based on a neural network model and disambiguation based on what we have called dynamic context disambiguation.
    [Show full text]
  • Knowledge-Driven Geospatial Location Resolution for Phylogeographic
    Bioinformatics, 31, 2015, i348–i356 doi: 10.1093/bioinformatics/btv259 ISMB/ECCB 2015 Knowledge-driven geospatial location resolution for phylogeographic models of virus migration Davy Weissenbacher1,*, Tasnia Tahsin1, Rachel Beard1,2, Mari Figaro1,2, Robert Rivera1, Matthew Scotch1,2 and Graciela Gonzalez1 1Department of Biomedical Informatics, Arizona State University, Scottsdale, AZ 85259, USA and 2Center for Environmental Security, Biodesign Institute, Arizona State University, Tempe, AZ 85287-5904, USA *To whom correspondence should be addressed. Abstract Summary: Diseases caused by zoonotic viruses (viruses transmittable between humans and ani- mals) are a major threat to public health throughout the world. By studying virus migration and mutation patterns, the field of phylogeography provides a valuable tool for improving their surveil- lance. A key component in phylogeographic analysis of zoonotic viruses involves identifying the specific locations of relevant viral sequences. This is usually accomplished by querying public data- bases such as GenBank and examining the geospatial metadata in the record. When sufficient detail is not available, a logical next step is for the researcher to conduct a manual survey of the corresponding published articles. Motivation: In this article, we present a system for detection and disambiguation of locations (topo- nym resolution) in full-text articles to automate the retrieval of sufficient metadata. Our system has been tested on a manually annotated corpus of journal articles related to phylogeography using inte- grated heuristics for location disambiguation including a distance heuristic, a population heuristic and a novel heuristic utilizing knowledge obtained from GenBank metadata (i.e. a ‘metadata heuristic’). Results: For detecting and disambiguating locations, our system performed best using the meta- data heuristic (0.54 Precision, 0.89 Recall and 0.68 F-score).
    [Show full text]
  • TAGGS: Grouping Tweets to Improve Global Geotagging for Disaster Response Jens De Bruijn1, Hans De Moel1, Brenden Jongman1,2, Jurjen Wagemaker3, Jeroen C.J.H
    Nat. Hazards Earth Syst. Sci. Discuss., https://doi.org/10.5194/nhess-2017-203 Manuscript under review for journal Nat. Hazards Earth Syst. Sci. Discussion started: 13 June 2017 c Author(s) 2017. CC BY 3.0 License. TAGGS: Grouping Tweets to Improve Global Geotagging for Disaster Response Jens de Bruijn1, Hans de Moel1, Brenden Jongman1,2, Jurjen Wagemaker3, Jeroen C.J.H. Aerts1 1Institute for Environmental Studies, VU University, Amsterdam, 1081HV, The Netherlands 5 2Global Facility for Disaster Reduction and Recovery, World Bank Group, Washington D.C., 20433, USA 3FloodTags, The Hague, 2511 BE, The Netherlands Correspondence to: Jens de Bruijn ([email protected]) Abstract. The availability of timely and accurate information about ongoing events is important for relief organizations seeking to effectively respond to disasters. Recently, social media platforms, and in particular Twitter, have gained traction as 10 a novel source of information on disaster events. Unfortunately, geographical information is rarely attached to tweets, which hinders the use of Twitter for geographical applications. As a solution, analyses of a tweet’s text, combined with an evaluation of its metadata, can help to increase the number of geo-located tweets. This paper describes a new algorithm (TAGGS), that georeferences tweets by using the spatial information of groups of tweets mentioning the same location. This technique results in a roughly twofold increase in the number of geo-located tweets as compared to existing methods. We applied this approach 15 to 35.1 million flood-related tweets in 12 languages, collected over 2.5 years. In the dataset, we found 11.6 million tweets mentioning one or more flood locations, which can be towns (6.9 million), provinces (3.3 million), or countries (2.2 million).
    [Show full text]
  • Structured Toponym Resolution Using Combined Hierarchical Place Categories∗
    In GIR’13: Proceedings of the 7th ACM SIGSPATIAL Workshop on Geographic Information Retrieval, Orlando, FL, November 2013 (GIR’13 Best Paper Award). Structured Toponym Resolution Using Combined Hierarchical Place Categories∗ Marco D. Adelfio Hanan Samet Center for Automation Research, Institute for Advanced Computer Studies Department of Computer Science, University of Maryland College Park, MD 20742 USA {marco, hjs}@cs.umd.edu ABSTRACT list or table is geotagged with the locations it references. To- ponym resolution and geotagging are common topics in cur- Determining geographic interpretations for place names, or rent research but the results can be inaccurate when place toponyms, involves resolving multiple types of ambiguity. names are not well-specified (that is, when place names are Place names commonly occur within lists and data tables, not followed by a geographic container, such as a country, whose authors frequently omit qualifications (such as city state, or province name). Our aim is to utilize the context of or state containers) for place names because they expect other places named within the table or list to disambiguate the meaning of individual place names to be obvious from place names that have multiple geographic interpretations. context. We present a novel technique for place name dis- ambiguation (also known as toponym resolution) that uses Geographic references are a very common component of Bayesian inference to assign categories to lists or tables data tables that can occur in both (1) the case where the containing place names, and then interprets individual to- primary entities of a table are geographic or (2) the case ponyms based on the most likely category assignments.
    [Show full text]
  • Toponym Resolution in Scientific Papers
    SemEval-2019 Task 12: Toponym Resolution in Scientific Papers Davy Weissenbachery, Arjun Maggez, Karen O’Connory, Matthew Scotchz, Graciela Gonzalez-Hernandezy yDBEI, The Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA zBiodesign Center for Environmental Health Engineering, Arizona State University, Tempe, AZ 85281, USA yfdweissen, karoc, [email protected] zfamaggera, [email protected] Abstract Disambiguation of toponyms is a more recent task (Leidner, 2007). We present the SemEval-2019 Task 12 which With the growth of the internet, the public adop- focuses on toponym resolution in scientific ar- ticles. Given an article from PubMed, the tion of smartphones equipped with Geographic In- task consists of detecting mentions of names formation Systems and the collaborative devel- of places, or toponyms, and mapping the opment of comprehensive maps and geographi- mentions to their corresponding entries in cal databases, toponym resolution has seen an im- GeoNames.org, a database of geospatial loca- portant gain of interest in the last two decades. tions. We proposed three subtasks. In Sub- Not only academic but also commercial and open task 1, we asked participants to detect all to- source toponym resolvers are now available. How- ponyms in an article. In Subtask 2, given to- ponym mentions as input, we asked partici- ever, their performance varies greatly when ap- pants to disambiguate them by linking them plied on corpora of different genres and domains to entries in GeoNames. In Subtask 3, we (Gritta et al., 2018). Toponym disambiguation asked participants to perform both the de- tackles ambiguities between different toponyms, tection and the disambiguation steps for all like Manchester, NH, USA vs.
    [Show full text]
  • A Survey on Geocoding: Algorithms and Datasets for Toponym Resolution
    A Survey on Geocoding: Algorithms and Datasets for Toponym Resolution Anonymous ACL submission Abstract survey and critical evaluation of the currently avail- 040 able datasets, evaluation metrics, and geocoding 041 001 Geocoding, the task of converting unstructured algorithms. Our contributions are: 042 002 text to structured spatial data, has recently seen 003 progress thanks to a variety of new datasets, • the first survey to review deep learning ap- 043 004 evaluation metrics, and machine-learning algo- proaches to geocoding 044 005 rithms. We provide a comprehensive survey • comprehensive coverage of geocoding sys- 045 006 to review, organize and analyze recent work tems, which increased by 50% in the last 4 046 007 on geocoding (also known as toponym resolu- 008 tion) where the text is matched to geospatial years 047 009 coordinates and/or ontologies. We summarize • comprehensive coverage of annotated geocod- 048 010 the findings of this research and suggest some ing datasets, which increased by 100% in the 049 011 promising directions for future work. last 4 years 050 012 1 Introduction 2 Background 051 013 Geocoding, also called toponym resolution or to- An early work on geocoding, Amitay et al.(2004), 052 014 ponym disambiguation, is the subtask of geopars- identifies two important types of ambiguity: A 053 015 ing that disambiguates place names in text. The place name may also have a non-geographic mean- 054 016 goal of geocoding is, given a textual mention of a ing, such as Turkey the country vs. turkey the ani- 055 017 location, to choose the corresponding geospatial co- mal, and two places may have the same name, such 056 018 ordinates, geospatial polygon, or entry in a geospa- as the San Jose in California and the San Jose in 057 019 tial database.
    [Show full text]
  • An Overview of Geotagging Geospatial Location
    An Overview of Geotagging Geospatial location Archival collections may include a huge amount of original documents which contain geospa- tial location (toponym) data. Scientists, researchers and archivists are aware of the value of this information, and various tools have been developed to extract this key data from archival records (Kemp, 2008). Geospatial identification metadata generally involves latitude and longitude coordi- nates, coded placenames and data sources, etc. (Hunter, 2012). Tagging Tagging, a natural result of social media and semantic web technologies, involves labelling content with arbitrarily selected tags. Geotags, which describe the geospatial information of the content, are one of the most common online tag types (Hu & Ge, 2008), (Intagorn , et al., 2010). Geotagging Geotagging refers to the process of adding geospatial identification metadata to various types of media such as e-books, narrative documents, web content, images and video and social media applications (Facebook, Twitter, Foursquare etc) (Hunter, 2012). With various social web tools and applications, users can add geospatial information to web content, photographs, audio and video. Using hardware integrated with Global Positioning System (GPS) receivers, this can be done automatically (Hu & Ge, 2008). 2 Social media tools with geo-tagging capabilities Del.icio.us1 A social bookmarking site that allows users to save and tag web pages and resources. CiteULike2 An online service to organise academic publications that allows users to tag academic papers and books. Twitter3 A real-time micro-blogging platform/information network that allows users all over the world to share and discover what is happening. Users can publish their geotag while tweeting. Flickr4 A photo-sharing service that allows users to store, tag and geotag their personal photos, maintain a network of contacts and tag others photos.
    [Show full text]
  • A Metadata Geoparsing System for Place Name Recognition and Resolution in Metadata Records
    A Metadata Geoparsing System for Place Name Recognition and Resolution in Metadata Records Nuno Freire, José Borbinha, Pável Calado, Bruno Martins IST / INESC-ID Apartado 13069, 1000-029 Lisboa, Portugal {nuno.freire, jlb, pavel.calado, bruno.g.martins}@ist.utl.pt ABSTRACT follow any data model at all, when it consists of natural language This paper describes an approach for performing recognition and texts (it was even estimated that 80% to 90% of business resolution of place names mentioned over the descriptive information may exist in those unstructured forms [1][2]). metadata records of typical digital libraries. Our approach exploits As society becomes more data oriented, much interest has arisen evidence provided by the existing structured attributes within the in these unstructured sources of information. This interest gave metadata records to support the place name recognition and origin to the research field of information extraction, which looks resolution, in order to achieve better results than by just using for automatic ways to create structured data from unstructured lexical evidence from the textual values of these attributes. In data sources [3]. An information extraction process selectively metadata records, lexical evidence is very often insufficient for structures and combines data that is found, explicitly stated or this task, since short sentences and simple expressions are implied, in texts or data sets with semi-structured schemas. The predominant. Our implementation uses a dictionary based final output of the extraction process will vary according to the technique for recognition of place names (with names provided by purpose, but typically it will consist in semantically richer data, Geonames), and machine learning for reasoning on the evidences which follows a more structured data model, and on which more and choosing a possible resolution candidate.
    [Show full text]
  • Geocoding Location Expressions in Twitter Messages: a Preference Learning Method
    JOURNAL OF SPATIAL INFORMATION SCIENCE Number 9 (2014), pp. 37–70 doi:10.5311/JOSIS.2014.9.170 RESEARCH ARTICLE Geocoding location expressions in Twitter messages: A preference learning method Wei Zhang and Judith Gelernter School of Computer Science, Carnegie Mellon University, Pittsburgh, USA Received: March 6, 2014; returned: May 12, 2014; revised: June 3, 2014; accepted: June 27, 2014. Abstract: Resolving location expressions in text to the correct physical location, also known as geocoding or grounding, is complicated by the fact that so many places around the world share the same name. 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, as may be the case among different metadata fields of a Twitter message. 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 the extracted expression. We evaluated our model using the F1 measure and compared it to similar algorithms. Our method achieved results higher than state-of-the-art competitors. Keywords: geocoding, toponym resolution, named entity disambiguation, geographic ref- erencing, geolocation, grounding, geographic information retrieval, Twitter 1 Introduction News tweet: “Van plows into Salvation Army store in Sydney.”1 The obvious guess is that this tweet refers to an incident in Australia. In fact, the tweet refers to an incident that happened on the Cape Breton Island in Nova Scotia, Canada.
    [Show full text]
  • Adaptive Geoparsing Method for Toponym Recognition and Resolution in Unstructured Text
    remote sensing Article Adaptive Geoparsing Method for Toponym Recognition and Resolution in Unstructured Text Edwin Aldana-Bobadilla 1,* , Alejandro Molina-Villegas 2 , Ivan Lopez-Arevalo 3 , Shanel Reyes-Palacios 3, Victor Muñiz-Sanchez 4 and Jean Arreola-Trapala 4 1 Conacyt-Centro de Investigación y de Estudios Avanzados del I.P.N. (Cinvestav), Victoria 87130, Mexico 2 Conacyt-Centro de Investigación en Ciencias de Información Geoespacial (Centrogeo), Mérida 97302, Mexico; [email protected] 3 Centro de Investigación y de Estudios Avanzados del I.P.N. Unidad Tamaulipas (Cinvestav Tamaulipas), Victoria 87130, Mexico; [email protected] (I.L.-A.); [email protected] (S.R.-P.) 4 Centro de Investigación en Matemáticas (Cimat), Monterrey 66628, Mexico; [email protected] (V.M.-S.); [email protected] (J.A.-T.) * Correspondence: [email protected] Received: 10 July 2020; Accepted: 13 September 2020; Published: 17 September 2020 Abstract: The automatic extraction of geospatial information is an important aspect of data mining. Computer systems capable of discovering geographic information from natural language involve a complex process called geoparsing, which includes two important tasks: geographic entity recognition and toponym resolution. The first task could be approached through a machine learning approach, in which case a model is trained to recognize a sequence of characters (words) corresponding to geographic entities. The second task consists of assigning such entities to their most likely coordinates. Frequently, the latter process involves solving referential ambiguities. In this paper, we propose an extensible geoparsing approach including geographic entity recognition based on a neural network model and disambiguation based on what we have called dynamic context disambiguation.
    [Show full text]
  • Toponym Resolution in Social Media
    Toponym Resolution in Social Media Neil Ireson and Fabio Ciravegna University of Sheffield, UK Abstract. Increasingly user-generated content is being utilised as a source of information, however each individual piece of content tends to contain low levels of information. In addition, such information tends to be informal and imperfect in nature; containing imprecise, subjec- tive, ambiguous expressions. However the content does not have to be interpreted in isolation as it is linked, either explicitly or implicitly, to a network of interrelated content; it may be grouped or tagged with similar content, comments may be added by other users or it may be re- lated to other content posted at the same time or by the same author or members of the author’s social network. This paper generally examines how ambiguous concepts within user-generated content can be assigned a specific/formal meaning by considering the expanding context of the information, i.e. other information contained within directly or indirectly related content, and specifically considers the issue of toponym resolution of locations. Keywords: Concept Disambiguation, Social networks, Information Extraction. 1 Introduction The growth in the use of social media for sharing content (text, images or video) to other individuals who can be close personal associates or random strangers, is staggering. If the latest statistics from Facebook1 are to be believed, around 7% of the world’s population are active users and they spend on average 40 minutes per day on this one site. Whilst the value of this User-Generated Content (UGC) is being realised, utilising the information it contains poses a number of challenges.
    [Show full text]
  • Toponym Resolution with Deep Neural Networks
    Toponym Resolution with Deep Neural Networks Ricardo Jorge Barreira Custódio Thesis to obtain the Master of Science Degree in Telecommunications and Computer Engineering Supervisors: Prof. Doutor Bruno Emanuel da Graça Martins Prof. Miguel Daiyen Carvalho Won Examination Committee Chairperson: Prof. Doutor Luís Eduardo Teixeira Rodrigues Supervisor: Prof. Doutor Bruno Emanuel da Graça Martins Members of the Committee: Prof. Doutor David Martins de Matos October 2017 Acknowledgements First, I would like to thank Professor Bruno Emanuel da Gra¸caMartins for his guidance during this last and challenging year. His knowledge and constant motivation were a major contribution to the work that we developed together. Second, I would also like to thank my family, in particular my parents and my sister, for their constant support and for giving me the opportunity to learn in such a distinguished institute as Instituto Superior Te´cnico. Finally, I have to thank all my friends and colleagues for the constant support during the hard, although also amazing, time spent at Instituto Superior Te´cnico. A special word of gratitude for my colleagues under the tutelage of Professor Bruno Martins, who went through this last year experience with me. The countless hours we spent together, helping and making fun of each other is something I will never forget. I share this achievement with you. Ricardo Jorge Barreira Cust´odio For my parents and sister, Resumo Resolu¸c~aode top´onimos,i.e., inferir as coordenadas geogr´aficasde uma string que representa o nome de um local, ´eum problema fundamental no contexto de v´ariasaplica¸c~oesrelacionadas com extra¸c~aode informa¸c~aogeogr´aficae das ci^enciasde informa¸c~aogeogr´afica.
    [Show full text]