An efficient node ordering method using the conditional frequency for the K2 algorithm.
Song KoDae-Won KimPublished in: Pattern Recognit. Lett. (2014)
Keyphrases
- dynamic programming
- high accuracy
- significant improvement
- tree structure
- experimental evaluation
- synthetic and real images
- cost function
- detection method
- preprocessing
- improved algorithm
- computational cost
- computationally efficient
- detection algorithm
- optimization method
- estimation algorithm
- objective function
- classification method
- highly efficient
- learning algorithm
- segmentation method
- optimization algorithm
- k means
- theoretical analysis
- clustering method
- computational efficiency
- experimental study
- mathematical model
- recognition algorithm
- high efficiency
- probabilistic model
- classification algorithm
- input data
- expectation maximization
- similarity measure
- neural network
- optimal path
- single pass
- selection algorithm
- parameter estimation
- em algorithm
- noisy data
- energy function
- support vector machine svm
- segmentation algorithm
- node selection
- prior information
- pairwise
- computational complexity
- convergence rate
- region of interest
- reconstruction method
- distance matrix