Classifiers

MUMT614


Introduction

  • Preprocessing
    • e.g.: segmentation, FFT, MFCC
  • Feature extraction (classical model)
    • e.g.: centorid, area
  • Feature selection
  • Classification
    • Trainig
    • Validation
      • holdout method
      • k-fold cross-validation
      • Bootstrapping

Classifiers (supervised)

  • Bayes classifier
  • Limitations
  • Support Vector Machines
  • Hidden Markov models
  • Non-parametric density estimation (distribution-free)
    • k-nearest neighbour
    • Neural networks

Clustering (unsupervised)

  • Hierachical methods (an example)
  • k-means (demo)
    • Gaussian mixture
  • Self Organizing Maps

Resources


Created: 2003.03.12 Modified: Ichiro Fujinaga
McGill Crest