4930 search results - page 192 / 986 » Learning minimal abstractions |
172
click to vote
FGR
2008
IEEE
16 years 1 months ago
2008
IEEE
Linear Discriminant Analysis (LDA) has been a popular method for feature extracting and face recognition. As a supervised method, it requires manually labeled samples for training...
178
click to vote
ICDM
2006
IEEE
16 years 18 days ago
2006
IEEE
Kernel methods have been shown to be very effective for applications requiring the modeling of structured objects. However kernels for structures usually are too computational dem...
173
click to vote
WAIM
2009
Springer
15 years 11 months ago
2009
Springer
In the k-nearest neighbor (KNN) classifier, nearest neighbors involve only labeled data. That makes it inappropriate for the data set that includes very few labeled data. In this ...
131
click to vote
KDD
2010
ACM
15 years 10 months ago
2010
ACM
Existing cost-sensitive learning methods work with unequal misclassification cost that is given by domain knowledge and appears as precise values. In many real-world applications,...
201
click to vote
ECCV
2010
Springer
15 years 8 months ago
2010
Springer
Many visual search and matching systems represent images using sparse sets of "visual words": descriptors that have been quantized by assignment to the best-matching symb...