Powered by OpenAIRE graph
Found an issue? Give us feedback
addClaim

Verortung geschichtswissenschaftlicher Quellen mittels Geoparsing

Locating historical sources using geoparsing
Authors: Novak, André;

Verortung geschichtswissenschaftlicher Quellen mittels Geoparsing

Abstract

„Historical GIS“, also die Anwendung von Methoden der Geoinformation in der Geschichts-wissenschaft, entstand im Zuge des spatial turns. Verbunden mit der Verbreitung der „Digital Humanities“ entwickeln sich so neue Arbeitsweisen und die Möglichkeit, auch ganz neue Fragen zu stellen. Da in der Geschichtswissenschaft vor allem mit textlichen Quellen gearbeitet wird, sollte es daher das Ziel sein, eine Arbeit mit diesen auch im Bereich des Historical GIS zu ermöglichen. Daher wurden „Qualitative GIS“-Ansätze gewählt, um dies zu verwirklichen. „Text Mining“ kann dabei helfen, räumliche Informationen aus Texten zu extra-hieren. Daher wurde ein „Geoparsing“-Workflow (Toponym Recognition + Resolution) entwickelt, welcher die Verortung geschichtswissenschaftlicher Quellen einfacher und automatisierbar macht. Dabei wurde das Online-Gazetteer „GeoNames“ für das Geocoding gewählt. Die Programmierung wurde mittels „Python“ und dem Natural-Language-Package „spaCy“ umgesetzt. Der Workflow wurde mit drei verschiedenen Quellentypen (Brief, Zeitung, Reisebericht) getestet. Auch wenn eine automatische Transkription einer Quelle mittels „OCR“ die Verortung beeinflusst, können solche Texte trotzdem zur Verortung herangezogen werden. Eine Evaluation der Ergebnisse zeigte, dass auch wenn der Workflow nur einen Teil der Ortsnamen richtig verorten konnte, trotzdem aussagekräftige Aussagen über den jeweiligen Text getroffen werden konnte, ohne diesen jemals genau gelesen haben zu müssen. Eine manuelle Nachbearbeitung kann dabei das Ergebnis deutlich verbessern. Trotzdem gibt es noch einige Optimierungsmöglichkeiten für zukünftige Arbeiten.

“Historical GIS”, that is the application of geographic information methods in historical studies, developed at the time of the spatial turn. In connection with “Digital Humanities”, new work methods and the possibility to answer completely new questions developed. Since in historical studies people mainly work with textual sources, the aim should be to work with them in the area of Historical GIS. For that reason, “Qualitative GIS”-approaches were chosen to realise this. “Text mining” can help to extract spatial information from texts. Therefore, a “geoparsing”-workflow (toponym recognition + resolution) was developed, which makes the localisation of historical sources easier and automatable. The online gazetteer “GeoNames” was chosen for geocoding. The programming was done in “Python” and the natural-language-processing package used was “spaCy”. The workflow was tested with three different types of sources (letter, newspaper, travel report). Even if an automatic transcription of a source using “OCR” influences the localisation, such texts can still be used for the “geoparsing”-process. An evaluation of the results showed that, although the workflow could only correctly locate parts of the place names, meaningful conclusions could still be drawn about the respective text without reading it closely. Manual post-processing can significantly improve the result. Nevertheless, there are still some optimisation options for future work.

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Upload OA version
Are you the author of this publication? Upload your Open Access version to Zenodo!
It’s fast and easy, just two clicks!