6533b820fe1ef96bd127a3de

RESEARCH PRODUCT

A new paradigm for pattern classification: Nearest Border Techniques

B. John OommenAlioune NgomLuis RuedaYifeng Li

subject

Class (set theory)Training setPattern ClassificationComputer sciencebusiness.industrySVMVDP::Mathematics and natural science: 400::Information and communication science: 420::Algorithms and computability theory: 422Centroid02 engineering and technology01 natural sciencesVDP::Mathematics and natural science: 400::Mathematics: 410::Analysis: 411Support vector machine010104 statistics & probabilityExperimental testingOutlier0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingArtificial intelligence0101 mathematics10. No inequalitySet (psychology)businessTest sampleBorder Identification

description

Published version of a chapter in the book: AI 2013: Advances in Artificial Intelligence. Also available from the publisher at: http://dx.doi.org/10.1007/978-3-319-03680-9_44 There are many paradigms for pattern classification. As opposed to these, this paper introduces a paradigm that has not been reported in the literature earlier, which we shall refer to as the Nearest Border (NB) paradigm. The philosophy for developing such a NB strategy is as follows: Given the training data set for each class, we shall first attempt to create borders for each individual class. After that, we advocate that testing is accomplished by assigning the test sample to the class whose border it lies closest to. This claim is actually counter-intuitive, because unlike the centroid or the median, these border samples are often “outliers” and are, really, the points that represent the class the least. However, we have formally proven this claim, and the theoretical results have been verified by rigorous experimental testing.

yearjournalcountryeditionlanguage
2013-01-01
10.1007/978-3-319-03680-9_44http://hdl.handle.net/11250/138014