IK-prototypes: Incremental Mixed Attribute Learning Based on K-Prototypes Algorithm, a New Method.
Siwar GorrabFahmi Ben RejabPublished in: ISDA (2020)
Keyphrases
- high accuracy
- preprocessing
- improved algorithm
- learning algorithm
- single pass
- cost function
- significant improvement
- detection algorithm
- optimization algorithm
- experimental evaluation
- dynamic programming
- clustering method
- recognition algorithm
- computational complexity
- detection method
- theoretical analysis
- synthetic and real images
- learning scheme
- incremental learning
- support vector machine svm
- segmentation algorithm
- k means
- matching algorithm
- classification algorithm
- similarity measure
- segmentation method
- computational cost
- learning process
- objective function
- selection algorithm
- prior knowledge
- estimation algorithm
- computationally efficient
- tree structure
- classification method
- optimization method
- energy function
- learning speed
- input data
- training process
- convergence rate
- incremental updating
- noisy data
- learning stage
- incremental version
- information entropy
- batch learning
- attribute set
- decision trees
- reconstruction method
- expectation maximization
- simulated annealing
- probabilistic model
- model free
- particle swarm optimization
- rough sets
- training set
- optimal solution
- delta bar delta