The Journal of Machine Learning Research
The Journal of Machine Learning Research
Image retrieval on large-scale image databases
Proceedings of the 6th ACM international conference on Image and video retrieval
Guest Editors' Introduction: Content Sharing
IEEE Pervasive Computing
An empirical comparison of tag clouds and tables
Proceedings of the 22nd Conference of the Computer-Human Interaction Special Interest Group of Australia on Computer-Human Interaction
eTACTS: A method for dynamically filtering clinical trial search results
Journal of Biomedical Informatics
Hi-index | 0.00 |
Flickr™, a web-based system for online photo management, has experienced rapid growth in popularity over the past several years, due to an increase in photo sharing as a social networking tool. Part of the excitement about Flickr is due to its use of tag clouds, which permit key words to be used as 'tags' associated with photos. Once tags are added to a photo, the picture can be retrieved for viewing by others based on the key words found the tag cloud associated with the photo. The research presented here examines the selection of words for inclusion in a tag cloud, and identifies the significance of the careful selection of tags. More frequently used words contribute to slower image retrieval, while less frequently used words, such as non-English terms, demonstrate a measurable increase in overall retrieval speed for the images associated with the tags. In addition, a personal tag cloud is built, integrated into Flickr, and shown to reduce the time taken for image retrieval by a factor of three, depending on the total number of images using the selected tag.