Prototype Selection for Nearest Neighbor Classification: Taxonomy and Empirical Study

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 34; no. 3; pp. 417 - 435
• Main Authors: Garcia, Salvador, Derrac, Joaquin, Cano, Jose Ramon, Herrera, Francisco
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.03.2012; IEEE Computer Society
• Subjects: Accuracy; Applied sciences; classification; Classification algorithms; Computer science; control theory; systems; condensation; Data processing. List processing. Character string processing; edition; Exact sciences and technology; Memory organisation. Data processing; nearest neighbor; Noise; Noise measurement; Prototype selection; Prototypes; Software; Taxonomy; Training; Performance evaluation; Capability index; Nearest neighbour; Data analysis; Taxonomy; Empirical method; edition; Data mining; classification; Condensation; Supervised classification; Recommendation; Prototype selection; Statistical test; nearest neighbor; Efficiency; Categorization
• ISSN: 0162-8828; 1939-3539; 2160-9292; 1939-3539
• DOI: 10.1109/TPAMI.2011.142

The nearest neighbor classifier is one of the most used and well-known techniques for performing recognition tasks. It has also demonstrated itself to be one of the most useful algorithms in data mining in spite of its simplicity. However, the nearest neighbor classifier suffers from several drawbac...