An improved K-medoids algorithm based on step increasing and optimizing medoids.
Donghua YuGuoJun LiuMaozu GuoXiaoyan LiuPublished in: Expert Syst. Appl. (2018)
Keyphrases
- experimental evaluation
- detection algorithm
- search space
- learning algorithm
- k means
- cost function
- single pass
- improved algorithm
- high accuracy
- preprocessing stage
- optimization algorithm
- segmentation algorithm
- worst case
- probabilistic model
- np hard
- computational cost
- preprocessing
- neural network
- similarity measure
- objective function
- selection algorithm
- parameter settings
- recognition algorithm
- convergence rate
- memory requirements
- genetic algorithm ga
- optimal solution
- lower bound
- theoretical analysis
- evolutionary algorithm
- significant improvement