Unsupervised Learning, Recognition, and Generation of Time-series Patterns Based on Self-Organizing Segmentation.
Shogo OkadaOsamu HasegawaPublished in: RO-MAN (2006)
Keyphrases
- unsupervised learning
- symbol recognition
- temporal patterns
- image segmentation
- object recognition
- neural network
- supervised learning
- image parsing
- recognition accuracy
- correct recognition
- segmentation algorithm
- medical images
- region growing
- level set
- numeral strings
- semi supervised
- multivariate time series
- pattern discovery
- automatic recognition
- recognition rate
- dimensionality reduction
- segmentation method
- sequential data
- recognition algorithm
- abnormal patterns
- pattern mining
- image analysis
- pattern recognition
- multiscale
- machine learning
- feature selection
- frequent patterns
- model selection
- surprising patterns
- character string
- stock data
- autoregressive
- non stationary
- edge detection
- data mining