TopoText is a digital mapping and text analysis software prototype was built by remixing features of open-source tools. After inputting a text in plain text (.txt) format, TopoText matches all unambiguous place names with geographical coordinates and displays them on a map interface using the Stanford Named Entity Recognition (NER) Tagger. Matching is carried out using the Google Maps Platform and placed onto a Google Maps Engine basemap. Coordinated with the map are text analysis and concordance tools that display the context in which the place names occur and allow manipulation of the content for analysis. Manipulation of text is carried out by collocating place-name occurrences and words that appear around them using Wordle. Users can specify the number of words to collocate around a place name, and can categorize them according to part of speech with the embedded Stanford Part-of-Speech (POS) Tagger, which extracts nouns, verbs, adjectives, or adverbs. These collocations can be localized to the specific passage in which the place name appears, or generated across the entire text by counting the most frequent word collocations around a selected place name.
The target users for this tool are researchers conducting a geospatial analysis of a work and exploring topics that occur around specific place names in or across texts. No technical background is required for using TopoText, nor does it require a steep learning curve; it was purposely designed to provide a quick and easy entry point from which to conduct a spatial analysis that could help guide further spatial investigations. In addition, users can upload any type of plain text in English, French, German, or Spanish; this option facilitates a multifocal approach and makes potential applications of the tool available to different disciplines.
Team
Randa El Khatib (project lead; University of Toronto), David Joseph Wrisley (NYU Abu Dhabi), Mohamad Jaber (Google), Julia El Zini, Bilal Abi Farraj, Shady Elbassuoni, Houda Nasser, Shadia Barada, Wassim El-Hajj, and Yasmin Kadah (American University of Beirut).
Keywords: geoparsing, experimental prototyping, word-place collocations, digital mapping
Randa El Khatib 