Generating geographical location descriptions with spatial templates: a salient toponym driven approach

ABSTRACT

Natural language descriptions of geographical locations are used frequently in daily life and there is a motivation to create systems that generate such descriptions automatically, for purposes such as documentation of where events have taken place, where a person is located, where photos were taken and where plants and animals are located. Typically location descriptions combine references to named geographical features with vague spatial relational terms, such as near, north of and at that relate locations to the features. Here we describe a system for generating location descriptions, that combines spatial templates, that model the applicability of different spatial relations relative to a reference location, with toponyms in the vicinity of the described location that are selected according to aspects of salience. The toponyms are retrieved from a gazetteer service based on OpenStreetMap for which we create a hierarchical feature classification scheme to facilitate selection of toponyms according to distinctiveness of their feature types and other aspects of salience. The advantages of the approach are demonstrated in a user study, relative to an existing state of the art system and to other baseline approaches that include manually created captions and the automated methods of two widely used photo captioning systems.

KEYWORDS:

• Spatial natural language
• referral expressions
• locative expressions
• salience
• landmarks
• spatial preposition applicability models
• vagueness
• natural language generation
• spatial templates
• gazetteers
• OpenStreetMap

Acknowledgments

Some of the material here is based upon work supported originally by the European Community in the TRIPOD project (FP6 cr n${ }^{\circ }$045335).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data and codes availability statement

The data and codes that support the findings of this study are available with the identifier https://doi.org/10.21954/ou.rd.12876938.v1.

Notes

1. http://www.geograph.org.uk/.

2. https://www.openstreetmap.org/.

3. https://www.flickr.com/.

4. https://www.eea.europa.eu/data-and-maps/data/clc-2006-raster-4.

5. The study was approved by the Research Ethics Committee of the Faculty of Arts and Sciences at Edgehill University, UK. Participants signed an electronic consent form that informed them of their freedom to withdraw from the study, anonymity in storage and use of the data, and giving permission for the research team members to use anonymised responses in research publications.

6. We define a preference as where the most popular description has at least three more votes than any other.

Additional information

Notes on contributors

Mark M. Hall

Mark M. Hall is a Lecturer in Computing & Communications at the Open University, UK. His research interests focus on the intersection between computation and the human user, particularly in information retrieval and the digital humanities. He is particularly interested in helping users explore large data-sets that they are unfamiliar with or where the users lack the expertise to know what they could find and how to find it. He is also exploring how to evaluate interfaces that support such open-ended exploration.

Christopher B. Jones

Christopher B. Jones is Professor of Geographical Information Systems at Cardiff University, having worked previously at the University of South Wales, the University of Cambridge, BP Exploration and the British Geological Survey. He studied geology at Imperial College and Bristol University and has a PhD on periodicities in fossil growth rhythms from the University of Newcastle upon Tyne. Current research focuses on geographical information retrieval, but earlier research on cartography led to the development of the Maplex cartographic name placement software that is now an ESRI product.