A Learning Pseudo Bayes Discriminant Method Based on Difference Distribution of Feature Vectors.
Hiroaki TakebeKoji KurokawaYutaka KatsuyamaSatoshi NaoiPublished in: Document Analysis Systems (2002)
Keyphrases
- feature vectors
- unsupervised learning
- preprocessing
- learning process
- neural network
- probabilistic model
- objective function
- detection method
- prior knowledge
- cost function
- classification accuracy
- model selection
- supervised learning
- dynamic programming
- bayesian methods
- online learning
- high accuracy
- probability distribution
- significant improvement
- computational complexity
- bayesian networks