Creating a Corpus of Geospatial Natural Language

Stock, Kristin; Pasley, Robert C.; Gardner, Zoë; Brindley, Paul J; Morley, Jeremy; Cialone, Claudia

doi:10.1007/978-3-319-01790-7_16

Cited by 12 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, we overview the usage of spatial semantics by various downstream tasks and killer applications including language grounding, navigation, self-driving cars, robotics (Tellex et al, 2011;Kollar et al, 2010), dialogue systems (Kelleher and Kruijff, 2006) and human machine interaction, and geographical information systems and knowledge graphs (Stock et al, 2013;Mai et al, 2020). Spatial semantics is very closely connected and relevant to visualization of natural language and grounding language into perception, central to dealing with configurations in the physical world and motivating a combination of vision and language for richer spatial understanding.…”

Section: Descriptionmentioning

confidence: 99%

Representation, Learning and Reasoning on Spatial Language for Downstream NLP Tasks

Kordjamshidi¹,

Pustejovsky²,

Moens³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Tutorial Abstracts

View full text Add to dashboard Cite

Section: Descriptionmentioning

confidence: 99%

Representation, Learning and Reasoning on Spatial Language for Downstream NLP Tasks

Kordjamshidi¹,

Pustejovsky²,

Moens³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Tutorial Abstracts

View full text Add to dashboard Cite

“…This trend of utilizing various kinds of volunteered or pseudo-volunteered spatial data from the Web shows the power of crowdsourcing (Rattenbury & Naaman, 2009), which provides a valuable alternative to the current data collection method in many research fields. Stock et al (2013) developed a set of geospatial syntactic templates for building a spatial natural language corpus with web data; this corpus reached a precision of 0.66 (i.e., 66% of the final corpus is evaluated as geospatial phrases). In this article, we lay out a framework for semi-automatically collecting and analyzing spatially stratified route directions sampled from the Web; then we present results that show statistically significant spatial patterns for the dominant use of relative over cardinal directions, and vice versa.…”

Section: Introductionmentioning

confidence: 99%

Exploring Regional Variation in Spatial Language Using Spatially Stratified Web-Sampled Route Direction Documents

Klippel

MacEachren

et al. 2014

Spatial Cognition & Computation

View full text Add to dashboard Cite

Spatial language, such as route directions, can be analyzed to shed light on how humans communicate and conceptualize spatial knowledge. This article details a computational linguistic approach using route directions to study regional variations in spatial language. We developed a web-sourcing approach to collect human-generated route direction documents on a geographical scale. Specifically, we built the Spatially strAtified Route Direction (SARD) Corpus through automated scraping, classifying, and georeferencing of route directions. Based on semantic categories of cardinal and relative direction terms, the analysis of the SARD Corpus reveals significant differences and patterns on both national (United States, United Kingdom, and Australia) and regional (contiguous U.S. states) levels. Combining computational linguistics and georeferencing approaches offers the potential for extending classic spatial linguistic studies.

show abstract

“…For example, it's possible to extract ZIP codes, addresses, and well-known landmarks automatically from documents. While the fields of GIR and NLP have contributed solutions that help users find information based on their interests, the possibility of automatically tracking spatiotemporal and semantic changes relating to events in text documents, and visualizing the extracted results using GIS is a new challenge (Janowicz et al 2012;Crooks et al 2013;Croitoru et al 2013;Stefanidis et al 2013;Wang and Stewart 2013;Stock et al 2013;Tsou and Leitner 2013;Tsou et al 2014;Wang and Stewart 2014). Extracting spatiotemporal and semantic information from a set of Web documents enables us to build a rich representation of the geographic knowledge described in text, capturing where, when, or what events occurred (Egenhofer 2002;Sankaranarayanan 2009;Larson and Shaw 2009;Joliveau et al 2011;Chasin et al 2014).…”

Section: Motivationmentioning

confidence: 99%

“…Geographic dynamics is an important topic in the field of GIScience, and involves understanding the fundamental characteristics of relevant forces and their underlying relationships in space and time (Yuan and Stewart Hornsby 2008;; Dodge et al 2008;Stewart Hornsby and Li 2009;Stewart and Wang 2013;Yuan 2013;Kwan et al 2014). In GIScience, modeling geographic information from web documents is a growing topic of research to which this dissertation makes an important contribution by focusing on handling the dynamics associated with spatiotemporal information in text (Egenhofer 2002;Larson and Shaw 2009;Joliveau et al 2011;Crooks et al 2013;Croitoru et al 2013;Stefanidis et al 2013;Wang and Stewart 2013;Stock et al 2013;Chasin et al 2014). The novel feature of this research is to extract the spatial and temporal information related to events based on the context of the text documents, and dynamically represent the results in a GIS environment, instead of exploring these two elements in isolation.…”

Section: Dissertation Outlinementioning

confidence: 99%

“…In the field of GIScience, researchers are growing increasingly interested in incorporating GIR techniques into their studies (Egenhofer 2002;Purves and Clough 2006;Purves et al 2007;Derungs et al 2012;Janowicz et al 2012;Crooks et al 2013;Croitoru et al 2013;Stefanidis et al 2013;Stock et al 2013;Chasin et al 2014). In GIScience, researchers are using GIR techniques to extract spatial information from a group of text documents related to routes, and represent paths on a map (Klippel et al 2008;Zhang et al 2012 to specific locations and times after 9/11.…”

Section: Related Work In Gir and Giscience Fieldsmentioning

confidence: 99%

See 1 more Smart Citation

Automated spatiotemporal and semantic information extraction for hazards

Wang¹

View full text Add to dashboard Cite

Geographical and Sustainability Science at University of Iowa for their help during my study period. I would like to thank my family and friends for all their supports and love during the 6 years for my Ph.D. study. Last but not least, I want to give my best thanks to my husband. He gives me a lot of supports and encouragement throughout my study.Without his supports, I could not achieve to the destination of the journey. Thank you to all of you.iv ABSTRACT This dissertation explores three research topics related to automated spatiotemporal and semantic information extraction about hazard events from Web news reports and other social media. The dissertation makes a unique contribution of bridging geographic information science, geographic information retrieval, and natural language processing.Geographic information retrieval and natural language processing techniques are applied to extract spatiotemporal and semantic information automatically from Web documents, to retrieve information about patterns of hazard events that are not explicitly described in the texts. Chapters 2, 3 and 4 can be regarded as three standalone journal papers. The research topics covered by the three chapters are related to each other, and are presented in a sequential way. Chapter 2 begins with an investigation of methods for automatically extracting spatial and temporal information about hazards from Web news reports. A set of rules is developed to combine the spatial and temporal information contained in the reports based on how this information is presented in text in order to capture the dynamics of hazard events (e.g., changes in event locations, new events occurring) as they occur over space and time. Chapter 3 presents an approach for retrieving semantic information about hazard events using ontologies and semantic gazetteers. With this work, information on the different kinds of events (e.g., impact, response, or recovery events) can be extracted as well as information about hazard events at different levels of detail. Using the methods presented in Chapter 2 and 3, an approach for automatically extracting spatial, temporal, and semantic information from tweets is discussed in Chapter 4. Four different elements of tweets are used for assigning appropriate spatial and temporal information to hazard events in tweets. Since tweets represent shorter, but more current information about hazards and how they are impacting a local area, key information about hazards can be retrieved through extracted spatiotemporal and semantic information from tweets.

show abstract

Creating a Corpus of Geospatial Natural Language

Cited by 12 publications

References 24 publications

Representation, Learning and Reasoning on Spatial Language for Downstream NLP Tasks

Representation, Learning and Reasoning on Spatial Language for Downstream NLP Tasks

Exploring Regional Variation in Spatial Language Using Spatially Stratified Web-Sampled Route Direction Documents

Automated spatiotemporal and semantic information extraction for hazards

Contact Info

Product

Resources

About