A non-parametric method to estimate the number of clusters.
André FujitaDaniel Yasumasa TakahashiAlexandre Galvão PatriotaPublished in: Comput. Stat. Data Anal. (2014)
Keyphrases
- high precision
- cost function
- estimation algorithm
- computational complexity
- high accuracy
- input data
- synthetic data
- experimental evaluation
- probabilistic model
- hierarchical clustering
- objective function
- preprocessing
- detection method
- computationally efficient
- classification accuracy
- clustering method
- em algorithm
- image registration
- data sets
- significant improvement
- prior knowledge
- pairwise
- similarity measure
- decision trees