6533b7d2fe1ef96bd125dfb5
RESEARCH PRODUCT
Learning vector quantization with alternative distance criteria
José Salvador SánchezFiliberto PlaF.j. Ferrisubject
Linde–Buzo–Gray algorithmLearning vector quantizationArtificial neural networkAdaptive algorithmbusiness.industryCodebookVector quantizationPattern recognitionDecision ruleMachine learningcomputer.software_genreComputingMethodologies_PATTERNRECOGNITIONLearning ruleArtificial intelligencebusinesscomputerMathematicsdescription
An adaptive algorithm for training of a nearest neighbour (NN) classifier is developed in this paper. This learning rule has some similarity to the well-known LVQ method, but uses the nearest centroid neighbourhood concept to estimate optimal locations of the codebook vectors. The aim of this approach is to improve the performance of the standard LVQ algorithms when using a very small codebook. The behaviour of the learning technique proposed here is experimentally compared to those of the plain k-NN decision rule and the LVQ algorithms.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2003-01-20 | Proceedings 10th International Conference on Image Analysis and Processing |