SBE-UKRI: Understanding imprecise space and time in narratives through qualitative representations, reasoning, and visualisation

Stanford University, Stanford CA

Investigators

Abstract

This project explores how Geographical Information Systems and related tools can identify, analyze and visualize qualitative senses of place alongside the quantitative spatial information that typify these systems. The project develops concepts and tools that enhance the ways that people integrate and interpret geographical data, including locations and associated sentiments, in large textual collections. The researchers examine places that are described in archival texts, including poetry and oral histories, and subsequently construct spatial representations of these textual geographies in the form of maps and visualizations. These goals are achieved by developing approaches to textual sources that allow imprecise and ambiguous spatial information to be used effectively by geospatial technologies that were initially designed for precisely measured spatial data. The project includes a team of historians, data scientists, and cartographers in collaboration with partner teams at research institutions in the United Kingdom. Additionally, the project supports graduate students and junior researchers as well as offering its results and tools to the wider scholarly community and heritage partners. This project builds upon two case studies comprising extensive textual corpora: a) poetic depictions of landscapes in the Lake District of England, an expressive and often metaphorical language of place; and b) survivor testimonies from the US Holocaust Museum, describing place and movement under extreme conditions and disorientation. These case studies allow the researchers to develop concepts and tools for complex and varied textual materials while enhancing the applications of the techniques with stakeholders working on cultural heritage projects. The approach is founded on the application of natural language processing and qualitative spatial and temporal reasoning to extract, locate, and contextualize imprecise information about places, which can subsequently be explored with advanced visual analytics and representation to test and communicate results. The project aims to show how these techniques can reveal not only where a place or event is located but also what it might feel like to be there and what it meant to particular historical actors. This proposal is awarded under the SBE-UKRI Lead Agency Opportunity. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.