Usually in natural language several amounts of geographic references are found, this comes with the common necessity of giving a context with time and location details. Considering these locational semantics, user queries may be satisfied in a more accurate way. In this work a geo-ontology is built for identifying geographic terms. A toponym disambiguation algorithm is proposed for assigning to a place name its corresponding location. The importance of geographic terms in documents is determined by means of a weighting strategy, that is also used to compare geographic contents and to provide a ranking of results. It is also provided a technique for combining a standard textual ranking and the obtained geographic ranking. Final results are evaluated using GeoCLEF test collection and baseline techniques.
Content
Author and article information
Contributors
Yisleidy Linares Zaila
Conference
Publication date:
September
2015
Publication date
(Print):
September
2015
Pages: 59-63
Affiliations
PhD. Student Computer Science and Engineering
University of Bologna
Mura Anteo Zamboni 7, BO 40138
Italy