TagNet: Using Soft Semantics to Search the Web

Geert Vanderhulst, Lieven Trappeniers

2013

Abstract

Semantic annotations are key to efficiently retrieve resources on the Web. On the one hand, ontologies underlying Web resources give rise to linked data. On the other hand, tagging has become increasingly popular to bring order in data across Web applications and social networks. While taxonomies and folksonomies serve the same purpose (i.e. classification), there is a large gap in semantics between uncontrolled keywords used for tagging and hierarchical concepts found in a taxonomy. In this paper we introduce sematags as light-weight `soft semantics' which aim to bridge the gap between `no semantics' and `hard semantics'. Sematags define aliases (synonyms) and isas (hypernyms) which overcome the typical issues conventional tags cope with such as ambiguity. Furthermore, we present TagNet, a framework that extracts sematags from lexicons and existing knowledge bases and exploits them to annotate, link and retrieve resources on the Web. We evaluate how soft semantics can be used to semi-automatically map tagged photos in Flickr on DBpedia concepts and vice versa.

Download


Paper Citation


in Harvard Style

Vanderhulst G. and Trappeniers L. (2013). TagNet: Using Soft Semantics to Search the Web . In Proceedings of the 9th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST, ISBN 978-989-8565-54-9, pages 179-188. DOI: 10.5220/0004325001790188

in Bibtex Style

@conference{webist13,
author={Geert Vanderhulst and Lieven Trappeniers},
title={TagNet: Using Soft Semantics to Search the Web},
booktitle={Proceedings of the 9th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,},
year={2013},
pages={179-188},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004325001790188},
isbn={978-989-8565-54-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Conference on Web Information Systems and Technologies - Volume 1: WEBIST,
TI - TagNet: Using Soft Semantics to Search the Web
SN - 978-989-8565-54-9
AU - Vanderhulst G.
AU - Trappeniers L.
PY - 2013
SP - 179
EP - 188
DO - 10.5220/0004325001790188