Experimental study on prototype optimisation algorithms for prototype-based classification in vector spaces

DOI10.1016/j.patcog.2006.04.005zbMath1097.68619MaRDI QIDQ2498675

Yanyan Li

Publication date: 16 August 2006

Published in: Pattern Recognition (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.patcog.2006.04.005

zbMATH Keywords

EM algorithm; prototype selection; dissimilarity representation; nearest neighbour rule; adaptive condensing; normal density based classifier

Mathematics Subject Classification ID

68T05: Learning and adaptive systems in artificial intelligence

68T10: Pattern recognition, speech recognition

Related Items

Supervised classification and mathematical optimization, Prototype selection for interpretable classification, A memetic algorithm for evolutionary prototype selection: A scaling up approach, Sparse conformal prediction for dissimilarity data, Generative models for similarity-based classification, Fast \(k\) most similar neighbor classifier for mixed data (tree \(k\)-MSN)

Uses Software

• UCI-ml
• PRTools
• LVQ_PAK

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Reduction techniques for instance-based learning algorithms
• Prototype selection for dissimilarity-based classifiers
• 10.1162/15324430260185592
• The Dissimilarity Representation for Pattern Recognition
• Statistical Properties of Error Estimators in Performance Assessment of Recognition Systems
• Finding Prototypes For Nearest Neighbor Classifiers
• Two Modifications of CNN
• Asymptotic Properties of Nearest Neighbor Rules Using Edited Data
• Dissimilarity representations allow for building good classifiers
• Structural, Syntactic, and Statistical Pattern Recognition
• Nearest neighbor pattern classification