AAAI
2006
14 years 1 months ago
2006
We present a novel framework for multi-label learning that explicitly addresses the challenge arising from the large number of classes and a small size of training data. The key a...