Abstract | ||
---|---|---|
TagMe! is a tagging and exploration front-end for Flickr images, which enables users to annotate specific areas of an image. i.e. users can attach tag assignments to a specific area within an image and further categorize the tag assignments. Additionally. TagMe! automatically maps tags and categories to DBpedia URIs to clearly define the meaning. In this work we discuss the differences between tags and categories and show how both facets can be applied to learn semantic relations between concepts referenced by tags and categories. We also expose the benefits of the visual (spatial) context of the tag assignments, with respect to ranking algorithms for search and retrieval of relevant items. We do so by analyzing metrics of size and position of the annotated areas. Finally, in our experiments we compare different strategies to realize semantic mappings and show that already lightweight approaches map tags and categories with high precisions (86.85% and 93.77% respectively). The TagMe! system is currently available at http://tagme.groupme.org. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-22810-0_9 | Lecture Notes in Business Information Processing |
Field | DocType | Volume |
Learning to rank,Categorization,World Wide Web,Information retrieval,Computer science,Spatial contextual awareness | Conference | 75 |
ISSN | Citations | PageRank |
1865-1348 | 0 | 0.34 |
References | Authors | |
10 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fabian Abel | 1 | 1187 | 62.22 |
Nicola Henze | 2 | 946 | 90.27 |
Ricardo Kawase | 3 | 100 | 9.99 |
Daniel Krause | 4 | 69 | 4.20 |
Patrick Siehndel | 5 | 126 | 15.69 |