Fair Overlap Number of Balls (Fair-ONB): A Data-Morphology-based Undersampling Method for Bias Reduction.
José Daniel Pascual-TrianaAlberto FernándezPaulo NovaisFrancisco HerreraPublished in: CoRR (2024)
Keyphrases
- synthetic data
- small number
- data sets
- test data
- input data
- computational complexity
- noisy data
- statistical analysis
- random samples
- original data
- statistical methods
- missing data
- preprocessing
- data analysis
- data structure
- prior knowledge
- data collection
- database
- high quality
- reduction method
- knowledge discovery
- objective function
- high accuracy
- correlation analysis
- em algorithm
- training samples
- clustering method
- significant improvement
- information loss
- data points
- spectral clustering
- prior information
- raw data
- detection method
- k means
- data mining
- microarray
- similarity measure
- edge detection
- image data
- classification accuracy