Semantic Historical Gazetteers and Related NLP and Corpus Linguistics Applications

Abstract

This issue of the Journal of Map & Geography Libraries presents extended contributions from the workshop “A Place for Places: Current Trends and Challenges in the Development and Use of Geo-historical Gazetteers,” held on July 11, 2016 in Kraków, Poland in conjunction with the Digital Humanities conference (DH2016). The workshop—endorsed by the GeoHumanities Special Interest Group (SIG) of the Alliance of Digital Humanities Organizations—was dedicated to an area of research that lies at the crossroads between geographic information science, corpus linguistics, natural language processing, and the semantic Web. It is undeniable that, in recent years many humanistic disciplines, such as History and Literary Studies, have undergone the effects of what has been termed a “digital turn,” and that this has led, among other things, to an increase in the application of quantitative approaches that strongly rely on natural language processing and corpus linguistics techniques for geographic information extraction and aggregation. Consequently, a plethora of projects has seen the light in recent years in which novel methods for analyzing and visualizing spatial-temporal phenomena described in documents from the past have been developed and put to use. Such methods depend heavily on algorithms for the automatic annotation of texts, such as Named Entity taggers and linkers, which have necessitated in turn the provision of reliable, domain-adapted sources of information. Understandably, then, a considerable amount of research has been devoted to the creation and use of semantic historical gazetteers, namely geographical data sets exposed as linked data describing places in their historical dimension. An exemplary project in this genre is the Pleiades gazetteer of the ancient world1 (Elliott and Gillies 2009). Built initially from cartographic resources describing the past, such as the Barrington Atlas, it has evolved into a digital semantic resource that is constantly being enlarged through the efforts of the community, and that has become in turn the basis of many other projects. Other noteworthy efforts include PastPlace2 (Southall, Mostern, and Berman 2011; Southall, von Luenen, and Aucott 2009) and the China Historical GIS.3