A new formalization of minimum classification error using a Parzen estimate of classification chance.
Erik McDermottShigeru KatagiriPublished in: ICASSP (2) (2003)
Keyphrases
- minimum classification error
- density estimation
- discriminative training
- classification accuracy
- support vector
- cross validation
- support vector machine
- feature space
- feature extraction
- decision trees
- pattern recognition
- feature vectors
- cost sensitive
- data sets
- supervised learning
- decision rules
- training samples
- non stationary
- knn
- benchmark datasets
- denoising
- hidden markov models
- training set
- image processing
- machine learning