Dealing with heterogeneous big data when geoparsing historical corpora

Rupp, C. J.; Rayson, Paul; Gregory, Ian; Hardie, Andrew; Joulain-Jay, Amelia; Hartmann, Daniel

doi:10.1109/bigdata.2014.7004457

Cited by 13 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conducting this process in an entirely automated manner was found not to be satisfactory for the complex place-names found in the CLDW (see Butler et al 2017). The process was enhanced using concordance geoparsing -where a small subset of the text is geoparsed, the results are checked, and any corrections fed into processing subsequent subsets (Rupp et al 2014) -and a considerable amount of manual checking.…”

Section: From Text To Gis Databasementioning

confidence: 99%

Mapping Landscapes in Transformation

Coomans¹,

Cattoor²,

Jonge³

2019

View full text Add to dashboard Cite

Section: From Text To Gis Databasementioning

confidence: 99%

Mapping Landscapes in Transformation

Coomans¹,

Cattoor²,

Jonge³

2019

View full text Add to dashboard Cite

“…These place-names then need to be matched to coordinates taken from a gazetteer to provide locations. When performed automatically, this process is known as geo-parsing [12,25]. At the end of this process, the coordinates can be used as spatial data and the cotext (the text immediately surrounding the place-name) can be used as attribute data, along with other information from either the text, its metadata or other datasets, such as gazetteers.…”

Section: Background 21 the Lake District Deep Mapping Projectmentioning

confidence: 99%

“…In previous research [24,25] we employed an Early Modern English variant spelling detector (VARD) to match historical to modern forms, and the DEEP 'Historical Gazetteer of England's Place-Names' 4 to improve coverage for the off-the-shelf geoparser. Based on this prior work we were able to determine that the Edinburgh Geoparser identified only 1277 of the 3718 place-name entities included in the Gold Standard subset.…”

Section: Gold Standard Corpusmentioning

confidence: 99%

A deeply annotated testbed for geographical text analysis

Rayson

Reinhold

Butler

et al. 2017

Proceedings of the 1st ACM SIGSPATIAL Workshop on Geospatial Humanities

Self Cite

View full text Add to dashboard Cite

This paper describes the development of an annotated corpus which forms a challenging testbed for geographical text analysis methods. This dataset, the Corpus of Lake District Writing (CLDW), consists of 80 manually digitised and annotated texts (comprising over 1.5 million word tokens). These texts were originally composed between 1622 and 1900, and they represent a range of different genres and authors. Collectively, the texts in the CLDW constitute an indicative sample of writing about the English Lake District during the early seventeenth century and the early twentieth century. The corpus is annotated more deeply than is currently possible with vanilla Named Entity Recognition, Disambiguation and geoparsing. This is especially true of the geographical information the corpus contains, since we have undertaken not only to link different historical and spelling variants of place-names, but also to identify and to differentiate geographical features such as waterfalls, woodlands, farms or inns. In addition, we illustrate the potential of the corpus as a gold standard by evaluating the results of three different NLP libraries and geoparsers on its contents. In the evaluation, the standard NER processing of the text by the different NLP libraries produces many false positive and false negative results, showing the strength of the gold standard.

show abstract

“…PNCs also provide us with a simpler, and often more accurate, way of geoparsing a text. By just geoparsing the text surrounding a search term we can restrict geoparsing to relevant parts of the corpus making the process quicker and easier to check for errors, a process known as concordance geoparsing (Rupp et al 2014). Figure 1a shows the PNCs created where toponyms are found within ten words of the search terms "tourist" and "tourists".…”

mentioning

confidence: 99%

Qualitative Geographies in Digital Texts: Representing Historical Spatial Identities in The Lake District

Smail¹,

Gregory²,

Taylor³

2019

IJHAC

View full text Add to dashboard Cite

Techniques for extracting place names (toponyms) from texts and using them to conduct analyses of the geographies within the texts are becoming reasonably well established. These are generally referred to as Geographical Text Analysis (GTA) and allow us to ask questions about the geographies within a corpus. The limitation of this approach is that the geographies that can be uncovered are solely associated with toponyms for which a coordinate-based location can be assigned. While this method is valuable, it is effectively a quantitative representation of the geographies associated with named places. Other representations of geography are ignored. To complement GTA, we need to develop techniques that are capable of representing the more qualitative representations of geography that are found within texts. Drawing on the Corpus of Lake District Writing, this paper presents some initial ideas about how this can be achieved, primarily by using techniques from corpus linguistics.

show abstract

Dealing with heterogeneous big data when geoparsing historical corpora

Cited by 13 publications

References 7 publications

Mapping Landscapes in Transformation

Mapping Landscapes in Transformation

A deeply annotated testbed for geographical text analysis

Qualitative Geographies in Digital Texts: Representing Historical Spatial Identities in The Lake District

Contact Info

Product

Resources

About