GeoAcademy: algorithm and platform for the automatic detection and location of geographic coordinates in scientific articles

Abstract

This document describes the GeoAcademy project, whose main objective is to automatically geolocate scientific articles, downloaded from general scientific databases such as Scopus or WoS. This geolocation is carried out on the content of the document, either through the capture of possible geographical coordinates that the document has, or toponyms that may appear in the document through an algorithm created for this purpose. In the methodology we explain the steps that have been taken in this project to create a sample database with articles that deal with Sierra Nevada (Spain) and the creation and design of the algorithm. The results show the technical data of the application of the algorithm on the database and its success rate, as well as a description of the platform created to graphically display the geolocated documents on a web map. Finally, in the discussion, we define the difficulties encountered, the possible bibliometric applications and its usefulness as a tool for viewing and retrieving information