DSpace UNL

RUN >
Instituto Superior de Estatística e Gestão de Informação (ISEGI) >
ISEGI - MSc Dissertations Geospatial Technologies (Erasmus-Mundus) >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10362/2548

Title: Enhancing information retrieval in folksonomies using ontology of place constructed from Gazetteer information
Authors: Sabrah, Rania Abd El Fattah Ahmed
Advisor: Kuhn, Werner
Bishr, Mohamed
Keywords: Semantics
Semantic web
Web 2.0
Folksonomy
Tagging systems
Gazetteer
Ontology
Ontology of place
Collective intelligence
Flickr
GeoNames
OWL
Issue Date: 9-Mar-2009
Series/Report no.: Master of Science in Geospatial Technologies;TGEO0017
Abstract: Folksonomy (from folk and taxonomy) is an approach to user metadata creation where users describe information objects with a free-form list of keywords (‘tags’). Folksonomy has have proved to be a useful information retrieval tool that support the emergence of “collective intelligence” or “bottom-up” light weight semantics. Since there are no guiding rules or restrictions on the users, folksonomy has some drawbacks and problems as lack of hierarchy, synonym control, and semantic precision. This research aims at enhancing information retrieval in folksonomy, particularly that of location information, by establishing explicit relationships between place name tags. To accomplish this, an automated approach is developed. The approach starts by retrieving tags from Flickr. The tags are then filtered to identify those that represent place names. Next, the gazetteer service that is a knowledge organization system for spatial information is used to query for the place names. The result of the search from the gazetteer and the feature types are used to construct an ontology of place. The ontology of place is formalized from place name concepts, where each place has a “Part-Of” relationship with its direct parent. The ontology is then formalized in OWL (Web Ontology Language). A search tool prototype is developed that extracts a place name and its parent name from the ontology and use them for searching in Flickr. The semantic richness added to Flickr search engine using our approach is tested and the results are evaluated.
Description: Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies
URI: http://hdl.handle.net/10362/2548
Appears in Collections:ISEGI - MSc Dissertations Geospatial Technologies (Erasmus-Mundus)

Files in This Item:

File Description SizeFormat
TGEO0017.pdf1.92 MBAdobe PDFView/Open
Statistics
View Statistics
FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpaceOrkut
Formato BibTex mendeley Endnote Logotipo do DeGóis 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Universidade Nova de Lisboa  - Statistics  - Feedback
Estamos no RCAAP Governo Português separator Ministério da Educação e Ciência   Fundação para a Ciência e a Tecnologia

Financiado por:

POS_C UE