Investigating behavioural and computational approaches for defining imprecise regions

Abstract

People often communicate with reference to informally agreed places, such as 'the city centre'. However, views of the spatial extent of such areas may vary and result in imprecise regions. We compare perceptions of Sheffield's City Centre from a street survey (with 61 participants) to spatial extents derived from various web-based sources. Such automated approaches have advantages of speed, cost and repeatability. Our results show that footprints derived from web sources are often in concordance with models derived from more labourintensive methods. There were, however, differences between some of the data sources, with those advertising/selling residential property diverging the most from the street survey data. Agreement between sources was measured by aggregating the web sources to identify locations of consensus.

Mapping a historic neighbourhood through user-generated content: the case of Alfama, Lisbon (Portugal)

“…In the literature, methods aimed at extracting boundaries and extents of urban regions from point-based data vary. Previous studies have applied kernel density estimation (KDE) to generate surfaces, extracting regions using different thresholds (Jones et al, 2008;Hollenstein and Purves, 2010;Brindley et al, 2018;Twaroch et al, 2019). Other methods include the α-shape algorithms (Arampatzis et al, 2006); fuzzy logic approaches; (Schockaert et al, 2005); concave hulls (Hu et al, 2015(Hu et al, , 2021; support-vector machines (Cunha and Martins, 2014); and Delaunay triangulation (Gao et al, 2017a).…”

“…Alike other historic neighbourhoods, Alfama is not bounded to administrative borders and hence its extent may vary according to people's activities, perceptions and experiences (Twaroch et al, 2019). Although neighbourhoods are spatially vague, fetching their boundaries is important not only for research purposes but also for practical applications such as geographic information retrieval in location-based services as well as outlining units for planning and execution of public policies (Shang et al, 2016;Brindley et al, 2018).…”

Mapping regions with partially defined boundaries

“…The complexities of boundedness in each of these four (or five) types of regions have been studied for decades, particularly the gradation zones in thematic regions (also known as categorical coverages, discrete fields, area‐class maps, or chorochromatic maps) that are manifestations of fuzziness in the categories they represent (Kronenfeld, 2005; Plewe, 1997), and vernacular regions such as urban neighborhoods that often have no boundaries defined at all (Davies, 2020; Twaroch et al., 2018). Cognitively, this does not seem to be a problem; it is possible to discuss at length regions such as “Downtown Phoenix” (vernacular), “A Desert” (thematic), and “Metro Phoenix” (functional) without worrying too much about their vague boundaries.…”

“…The closest thread of research involves the estimation of the extent of vernacular regions (such as urban neighborhoods) referenced as place names in text, as summarized by Twaroch et al. (2018) and others. The most common purpose of this research is to respond to internet queries involving vague places and vague spatial relationships (e.g., “list the hotels near Downtown Phoenix”).…”

Geospatial Semantics Analysis of the Qinghai–Tibetan Plateau Based on Microblog Short Texts

“…Egenhofer clarified three types of geospatial semantics: semantics of geospatial entity classes, semantics of spatial predicates, and semantics of geospatial names [7], which have been widely researched [8][9][10][11][12][13][14][15][16]. As well as the classes, names and spatial relations of geographical entities having meanings, the geographical entities themselves also have meanings; this is known as place semantics [17].…”