Information retrieval and visualization of geographic places coming from online newspapers

Abstract

During the last few years, the amount of online newspapers has reached reasonable dimension for many cities in the world. Such enriched information can now support semantic analysis of space, particularly in which respects to what places are covered all over the world news and what happens there. We present a system to automatically extract places of the most important newspapers covers in order to visualize them in a World projection. The newspapers analysed are the New York Times (USA), Le Monde (France), Morgen (Belgium), ABC (Spain), Il Giornale (Italy), Haberturk (Turkey), Tages Anzeiger (Swiss), Dagens Hyheter (Sweden), Telegraph (United Kingdom), Jornal de Noticias (Portugal), AD (Netherlands), Irish Examiner (Ireland), Gulf News (Arab Emirates), The Times of India (India), Haaretx (Israel), Sydney Morning Herald (Australia), Asahi (Japan), Reforma (Mexico), La Razón (Argentina), El Comercio (Peru), The Times (South African) and The Mercury (South African). To achieve this we use a Geoparsing Web Service to extract geographic coordinates from the online covers. Once geographic coordinates are extracted by using eXtensible Markup Language (XML) we draw the geo-positions and link the documents into a map image in order to visualize the places and the news concerning each place. It is possible, also, to search the title of the news according to a concept selected by the user.