Combining nearest neighbor data description and structural risk minimization for one-class classification
Identifieur interne : 000A90 ( Main/Exploration ); précédent : 000A89; suivant : 000A91Combining nearest neighbor data description and structural risk minimization for one-class classification
Auteurs : George G. Cabral [Brésil] ; Adriano L. I. Oliveira [Brésil] ; Carlos B. G. Cahu [Brésil]Source :
- Neural computing & applications : (Print) [ 0941-0643 ] ; 2009.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Classification, Prototype.
English descriptors
- KwdEn :
Abstract
One-class classification is an important problem with applications in several different areas such as novelty detection, anomaly detection, outlier detection and machine monitoring. In this paper, we propose two novel methods for one-class classification, referred to as NNDDSRM and kNNDDSRM. The methods are based on the principle of structural risk minimization and the nearest neighbor data description (NNDD) one-class classifier. Experiments carried out using both artificial and real-world datasets show that the proposed methods are able to significantly reduce the number of stored prototypes in comparison to NNDD. The experimental results also show that the proposed methods outperformed NNDD-in terms of the area under the receiver operating characteristic (ROC) curve-on four of the five datasets considered in the experiments and had a similar performance on the remaining one.
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream PascalFrancis, to step Corpus: 000238
- to stream PascalFrancis, to step Curation: 000541
- to stream PascalFrancis, to step Checkpoint: 000207
- to stream Main, to step Merge: 000B01
- to stream Main, to step Curation: 000A90
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Combining nearest neighbor data description and structural risk minimization for one-class classification</title><author><name sortKey="Cabral, George G" sort="Cabral, George G" uniqKey="Cabral G" first="George G." last="Cabral">George G. Cabral</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computing Systems, Polytechnic School of Engineering, Pemambuco State University, Rua Benfica, 455, Madalena</s1><s2>Recife, PE 50750-410</s2><s3>BRA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Brésil</country><wicri:noRegion>Recife, PE 50750-410</wicri:noRegion></affiliation></author><author><name sortKey="Oliveira, Adriano L I" sort="Oliveira, Adriano L I" uniqKey="Oliveira A" first="Adriano L. I." last="Oliveira">Adriano L. I. Oliveira</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computing Systems, Polytechnic School of Engineering, Pemambuco State University, Rua Benfica, 455, Madalena</s1><s2>Recife, PE 50750-410</s2><s3>BRA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Brésil</country><wicri:noRegion>Recife, PE 50750-410</wicri:noRegion></affiliation></author><author><name sortKey="Cahu, Carlos B G" sort="Cahu, Carlos B G" uniqKey="Cahu C" first="Carlos B. G." last="Cahu">Carlos B. G. Cahu</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computing Systems, Polytechnic School of Engineering, Pemambuco State University, Rua Benfica, 455, Madalena</s1><s2>Recife, PE 50750-410</s2><s3>BRA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Brésil</country><wicri:noRegion>Recife, PE 50750-410</wicri:noRegion></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">INIST</idno><idno type="inist">09-0116001</idno><date when="2009">2009</date><idno type="stanalyst">PASCAL 09-0116001 INIST</idno><idno type="RBID">Pascal:09-0116001</idno><idno type="wicri:Area/PascalFrancis/Corpus">000238</idno><idno type="wicri:Area/PascalFrancis/Curation">000541</idno><idno type="wicri:Area/PascalFrancis/Checkpoint">000207</idno><idno type="wicri:doubleKey">0941-0643:2009:Cabral G:combining:nearest:neighbor</idno><idno type="wicri:Area/Main/Merge">000B01</idno><idno type="wicri:Area/Main/Curation">000A90</idno><idno type="wicri:Area/Main/Exploration">000A90</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Combining nearest neighbor data description and structural risk minimization for one-class classification</title><author><name sortKey="Cabral, George G" sort="Cabral, George G" uniqKey="Cabral G" first="George G." last="Cabral">George G. Cabral</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computing Systems, Polytechnic School of Engineering, Pemambuco State University, Rua Benfica, 455, Madalena</s1><s2>Recife, PE 50750-410</s2><s3>BRA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Brésil</country><wicri:noRegion>Recife, PE 50750-410</wicri:noRegion></affiliation></author><author><name sortKey="Oliveira, Adriano L I" sort="Oliveira, Adriano L I" uniqKey="Oliveira A" first="Adriano L. I." last="Oliveira">Adriano L. I. Oliveira</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computing Systems, Polytechnic School of Engineering, Pemambuco State University, Rua Benfica, 455, Madalena</s1><s2>Recife, PE 50750-410</s2><s3>BRA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Brésil</country><wicri:noRegion>Recife, PE 50750-410</wicri:noRegion></affiliation></author><author><name sortKey="Cahu, Carlos B G" sort="Cahu, Carlos B G" uniqKey="Cahu C" first="Carlos B. G." last="Cahu">Carlos B. G. Cahu</name><affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Department of Computing Systems, Polytechnic School of Engineering, Pemambuco State University, Rua Benfica, 455, Madalena</s1><s2>Recife, PE 50750-410</s2><s3>BRA</s3><sZ>1 aut.</sZ><sZ>2 aut.</sZ><sZ>3 aut.</sZ></inist:fA14><country>Brésil</country><wicri:noRegion>Recife, PE 50750-410</wicri:noRegion></affiliation></author></analytic><series><title level="j" type="main">Neural computing & applications : (Print)</title><title level="j" type="abbreviated">Neural comput. appl. : (Print)</title><idno type="ISSN">0941-0643</idno><imprint><date when="2009">2009</date></imprint></series></biblStruct></sourceDesc><seriesStmt><title level="j" type="main">Neural computing & applications : (Print)</title><title level="j" type="abbreviated">Neural comput. appl. : (Print)</title><idno type="ISSN">0941-0643</idno></seriesStmt></fileDesc><profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Anomaly</term><term>Application</term><term>Classification</term><term>Classifier</term><term>Description</term><term>Detection</term><term>Experimental result</term><term>Minimization</term><term>Monitoring</term><term>Neural computation</term><term>Neural network</term><term>Outlier</term><term>Prototype</term><term>Receiver</term><term>Receiver operating characteristic curves</term><term>Roc curve</term></keywords><keywords scheme="Pascal" xml:lang="fr"><term>Description</term><term>Minimisation</term><term>Classification</term><term>Détection</term><term>Anomalie</term><term>Observation aberrante</term><term>Monitorage</term><term>Classificateur</term><term>Prototype</term><term>Résultat expérimental</term><term>Récepteur</term><term>Méthode ROC</term><term>Application</term><term>Réseau neuronal</term><term>Calcul neuronal</term><term>62H30</term><term>OCR</term><term>Courbe Roc</term></keywords><keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Classification</term><term>Prototype</term></keywords></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">One-class classification is an important problem with applications in several different areas such as novelty detection, anomaly detection, outlier detection and machine monitoring. In this paper, we propose two novel methods for one-class classification, referred to as NNDDSRM and kNNDDSRM. The methods are based on the principle of structural risk minimization and the nearest neighbor data description (NNDD) one-class classifier. Experiments carried out using both artificial and real-world datasets show that the proposed methods are able to significantly reduce the number of stored prototypes in comparison to NNDD. The experimental results also show that the proposed methods outperformed NNDD-in terms of the area under the receiver operating characteristic (ROC) curve-on four of the five datasets considered in the experiments and had a similar performance on the remaining one.</div></front></TEI><affiliations><list><country><li>Brésil</li></country></list><tree><country name="Brésil"><noRegion><name sortKey="Cabral, George G" sort="Cabral, George G" uniqKey="Cabral G" first="George G." last="Cabral">George G. Cabral</name></noRegion><name sortKey="Cahu, Carlos B G" sort="Cahu, Carlos B G" uniqKey="Cahu C" first="Carlos B. G." last="Cahu">Carlos B. G. Cahu</name><name sortKey="Oliveira, Adriano L I" sort="Oliveira, Adriano L I" uniqKey="Oliveira A" first="Adriano L. I." last="Oliveira">Adriano L. I. Oliveira</name></country></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000A90 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000A90 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= Pascal:09-0116001
|texte= Combining nearest neighbor data description and structural risk minimization for one-class classification
}}
| This area was generated with Dilib version V0.6.32. |  |