Discovering Location Indicators of Toponyms from News to Improve Gazetteer-Based Geo-Referencing

Cleber GouvêaStanley LohLuís Fernando Fortes GarciaEvandro Brasil da FonsecaIgor Wendt

This paper presents an approach that identifies Location Indicatorsrelated to geographical locations, by analyzing texts of news published in theWeb. The goal is to semi-automatically create Gazetteers with the identifiedrelations and then perform geo-referencing of news. Location Indicatorsinclude non-geographical entities that are dynamic and may change along thetime. The use of news published in the Web is a useful way to discoverLocation Indicators, covering a great number of locations and maintainingdetailed information about each location. Different training news corpora arecompared for the creation of Gazetteers and evaluated by their ability tocorrectly identify cities in texts of news.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: