118 search results - page 20 / 24 » Unlabeled data improves word prediction |
ICCV
2009
IEEE
16 years 4 months ago
2009
IEEE
Image auto-annotation is an important open problem in
computer vision. For this task we propose TagProp, a discriminatively
trained nearest neighbor model. Tags of test
images a...
COMCOM
2008
14 years 11 months ago
2008
A user's profile, for the purpose of location management in a personal communication service (PCS) network, is formalized as a subgraph of the network graph. This subgraph, t...
74
Voted
EMNLP
2010
14 years 9 months ago
2010
We examine effects that empty categories have on machine translation. Empty categories are elements in parse trees that lack corresponding overt surface forms (words) such as drop...
118
click to vote
KDD
2008
ACM
16 years 2 days ago
2008
ACM
Document classification presents difficult challenges due to the sparsity and the high dimensionality of text data, and to the complex semantics of the natural language. The tradi...
105
click to vote
ICWN
2008
15 years 1 months ago
2008
: Problem statement: For a sensor network comprising autonomous and self-organizing data sources, efficient similarity-based search for semantic-rich resources (such as video data)...