: a new support vector method for optimizing partial AUC based on a tight convex upper bound.
Harikrishna NarasimhanShivani AgarwalPublished in: KDD (2013)
Keyphrases
- upper bound
- error rate
- detection method
- significant improvement
- classification accuracy
- objective function
- support vector
- lower bound
- cross validation
- computational complexity
- clustering algorithm
- high accuracy
- preprocessing
- dynamic programming
- multi class
- cost function
- feature set
- model selection
- naive bayes
- clustering method
- pairwise
- feature selection
- convex hull
- roc curve
- decision function