Increased classification accuracy and speedup through pair-wise feature selection for support vector machines.
Kurt KramerDmitry B. GoldgofLawrence O. HallAndrew RemsenPublished in: CIDM (2011)
Keyphrases
- classification accuracy
- pairwise
- feature set
- feature selection
- naive bayes
- feature subset
- orders of magnitude
- feature space
- feature vectors
- multiple features
- feature representation
- multi class
- pattern recognition
- semi supervised
- support vector
- markov random field
- image sequences
- image segmentation
- computer vision
- genetic algorithm
- real time