The University of Sheffield
Department of Computer Science

Alexandru Florea Undergraduate Dissertation 2014/15

Mapping spatial references in text using Google Maps

Supervised by R.Gaizauskas

Abstract

We live in a time where geospatial information has become such an integral part of our lives that we do not give it much thought anymore. We 'check in' on our favourite social media platform using location detection, we plan our journey home with Google Maps and we are offered  "bird's eye" views to ease our understanding of news from areas of the world we might not be familiar with, all happening naturally on a daily basis. However, extensive research has been conducted so as to enable us to automate this process through the use of geoparsers, systems used for the identification of spatial references in text.

The aim of this project is to create a browser-based interface for an existing geoparsing system. It is meant to be used in conjunction with a web page so as to view the locations it mentions displayed on a map and have them highlighted in the text. While geospatial references can be found in any type of text, our focus is mainly on news articles.

After conducting considerable research we have chosen to implement the system as a Google Chrome browser extension that interfaces with a locally hosted CLIFF geoparsing server and displays the locations using Google Maps. We give background to the field of georeferencing, outline the main interface design principles and we motivate our choice of geoparser, interface type and map. The full implementation process and evaluation results are described before giving some suggestions for further improvement.