Abstract
High feature dimensionality of realistic datasets adversely affects the recognition accuracy of nearest neighbor (NN) classifiers. To address this issue, we introduce a nearest feature classifier that shifts the NN concept from the global-decision level to the level of individual features. Performance comparisons with 12 instance-based classifiers on 13 benchmark University of California Irvine classification datasets show average improvements of 6 and 3.5% in recognition accuracy and area under curve performance measures, respectively. The statistical significance of the observed performance improvements is verified by the Friedman test and by the post hoc Bonferroni-Dunn test. In addition, the application of the classifier is demonstrated on face recognition databases, a character recognition database and medical diagnosis problems for binary and multi-class diagnosis on databases including morphological and gene expression features.
| Original language | English |
|---|---|
| Pages (from-to) | 1072-1087 |
| Number of pages | 16 |
| Journal | Computer Journal |
| Volume | 55 |
| Issue number | 9 |
| DOIs | |
| Publication status | Published - Sep 2012 |
| Externally published | Yes |
Fingerprint
Keywords
- classification
- local features
- local ranking
- nearest neighbors
ASJC Scopus subject areas
- Computer Science(all)
Cite this
Nearest neighbor classifier based on nearest feature decisions. / James, Alex Pappachen; Dimitrijev, Sima.
In: Computer Journal, Vol. 55, No. 9, 09.2012, p. 1072-1087.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Nearest neighbor classifier based on nearest feature decisions
AU - James, Alex Pappachen
AU - Dimitrijev, Sima
PY - 2012/9
Y1 - 2012/9
N2 - High feature dimensionality of realistic datasets adversely affects the recognition accuracy of nearest neighbor (NN) classifiers. To address this issue, we introduce a nearest feature classifier that shifts the NN concept from the global-decision level to the level of individual features. Performance comparisons with 12 instance-based classifiers on 13 benchmark University of California Irvine classification datasets show average improvements of 6 and 3.5% in recognition accuracy and area under curve performance measures, respectively. The statistical significance of the observed performance improvements is verified by the Friedman test and by the post hoc Bonferroni-Dunn test. In addition, the application of the classifier is demonstrated on face recognition databases, a character recognition database and medical diagnosis problems for binary and multi-class diagnosis on databases including morphological and gene expression features.
AB - High feature dimensionality of realistic datasets adversely affects the recognition accuracy of nearest neighbor (NN) classifiers. To address this issue, we introduce a nearest feature classifier that shifts the NN concept from the global-decision level to the level of individual features. Performance comparisons with 12 instance-based classifiers on 13 benchmark University of California Irvine classification datasets show average improvements of 6 and 3.5% in recognition accuracy and area under curve performance measures, respectively. The statistical significance of the observed performance improvements is verified by the Friedman test and by the post hoc Bonferroni-Dunn test. In addition, the application of the classifier is demonstrated on face recognition databases, a character recognition database and medical diagnosis problems for binary and multi-class diagnosis on databases including morphological and gene expression features.
KW - classification
KW - local features
KW - local ranking
KW - nearest neighbors
UR - http://www.scopus.com/inward/record.url?scp=84865779001&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84865779001&partnerID=8YFLogxK
U2 - 10.1093/comjnl/bxs001
DO - 10.1093/comjnl/bxs001
M3 - Article
VL - 55
SP - 1072
EP - 1087
JO - Computer Journal
JF - Computer Journal
SN - 0010-4620
IS - 9
ER -