2011 | OriginalPaper | Buchkapitel
Geoparsing of Czech RSS News and Evaluation of Its Spatial Distribution
verfasst von : Jiří Horák, Pavel Belaj, Igor Ivan, Peter Nemec, Jiří Ardielli, Jan Růžička
Erschienen in: Semantic Methods for Knowledge Management and Communication
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Geoparsing assigns geographic identifiers to textual words and phrases in documents. The specific problem is how to apply geoparsing in languages where changes of word termination occur. An appropriate method requires a flexible solution reflecting different strategies and priorities. Sixteen Czech RSS news channels were evaluated according to ten criteria. Three selected RSS channels were monitored for more than two years. The applied geoparsing included successive steps of different filters’ application and utilized the generation of different grammatical cases for recognized entities. Various problems with geographical names are classified and documented. The quality assessment shows satisfactory results namely for identification of names in domiciles (94%). The pessimistic strategy is applied to analyze a geographical balance of news distribution. The results show significant differences between distribution of news in monitored channels and document a high concentration of cultural and national news in several locations.