A class-dependent weighted dissimilarity measure for nearest neighbor classification problems

Pattern Recognition Letters

Volumn 21, Issue 12, 2000, Pages 1027-1036

  Paredes, Roberto ^a   Vidal, Enrique ^a  

^a INSTITUTO TECNOLÓGICO DE INFORMÁTICA   (Spain)

Author keywords

Fractional programming; Iterative optimization; Nearest neighbour classification; Weighted dissimilarity measures

Indexed keywords

DATA REDUCTION; DATA STRUCTURES; ITERATIVE METHODS; MATHEMATICAL PROGRAMMING; VECTORS;

FRACTIONAL PROGRAMMING; NEAREST NEIGHBOUR CLASSIFICATION; WEIGHTED DISSIMILARITY MEASURE;

PATTERN RECOGNITION;

CLASSIFICATION; NEAREST NEIGHBOR ANALYSIS; OPTIMIZATION;

EID: 0034319192     PISSN: 01678655     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0167-8655(00)00064-7     Document Type: Article

References (21)

1. Nearest neighbor pattern classification
2. A parametrically defined nearest neighbour measure
3. An optimal global nearest neighbour metric
4. The 2-nn rule for more accurate nn risk estimation
5. An alternative nearest neighbour classification scheme
6. The multi-class metric problem in nearest neighbour discrimination rules
7. A nearest neighbor weighted measure in classification problems
8. Weighting prototypes. A new editing approach
9. Small sample effects in statistical pattern recognition: Recommendations for practitioners
10. A new nearest neighbor distance measure
11. An optimal distance measure for nearest neighbour classification
12. Modeling cognitive development on balance scale phenomena
13. Classification of radar returns from the ionosphere using neural networks
14. A generalization of the k-nn rule
15. Gradient descent learning of nearest neighbor classifiers with outlier rejection
16. Fast computation of normalized edit distances