Convex and nonconvex nonparametric ...
Type de document :
Compte-rendu et recension critique d'ouvrage
Titre :
Convex and nonconvex nonparametric frontier-based classification methods for anomaly detection
Auteur(s) :
Jin, Qianying [Auteur]
Nanjing University of Aeronautics and Astronautics [Nanjing] [NUAA]
Kerstens, Kristiaan [Auteur]
Lille économie management - UMR 9221 [LEM]
van de Woestyne, Ignace [Auteur]
Catholic University of Leuven = Katholieke Universiteit Leuven [KU Leuven]
Nanjing University of Aeronautics and Astronautics [Nanjing] [NUAA]
Kerstens, Kristiaan [Auteur]

Lille économie management - UMR 9221 [LEM]
van de Woestyne, Ignace [Auteur]
Catholic University of Leuven = Katholieke Universiteit Leuven [KU Leuven]
Titre de la revue :
OR Spectrum
Éditeur :
Springer Verlag
Date de publication :
2024-03-29
ISSN :
0171-6468
Mot(s)-clé(s) en anglais :
Nonparametric frontier
Convex
Nonconvex
Anomaly detection
Convex
Nonconvex
Anomaly detection
Discipline(s) HAL :
Sciences de l'Homme et Société/Economies et finances
Résumé en anglais : [en]
Efective methods for determining the boundary of the normal class are very useful for detecting anomalies in commercial or security applications—a problem known as anomaly detection. This contribution proposes a nonparametric ...
Lire la suite >Efective methods for determining the boundary of the normal class are very useful for detecting anomalies in commercial or security applications—a problem known as anomaly detection. This contribution proposes a nonparametric frontier-based classifcation (NPFC) method for anomaly detection. By relaxing the commonly used convexity assumption in the literature, a nonconvex-NPFC method is constructed and the nonconvex nonparametric frontier turns out to provide a more conservative boundary enveloping the normal class. By refecting on the monotonic relation between the characteristic variables and the membership, the proposed NPFC method is in a more general form since both input-like and outputlike characteristic variables are incorporated. In addition, by allowing some of the training observations to be misclassifed, the convex- and nonconvex-NPFC methods are extended from a hard nonparametric frontier to a soft one, which also provides a more conservative boundary enclosing the normal class. Both simulation studies and a real-life data set are used to evaluate and compare the proposed NPFC methods to some well-established methods in the literature. The results show that the proposed NPFC methods have competitive classifcation performance and have consistent advantages in detecting abnormal samples, especially the nonconvex-NPFC methods.Lire moins >
Lire la suite >Efective methods for determining the boundary of the normal class are very useful for detecting anomalies in commercial or security applications—a problem known as anomaly detection. This contribution proposes a nonparametric frontier-based classifcation (NPFC) method for anomaly detection. By relaxing the commonly used convexity assumption in the literature, a nonconvex-NPFC method is constructed and the nonconvex nonparametric frontier turns out to provide a more conservative boundary enveloping the normal class. By refecting on the monotonic relation between the characteristic variables and the membership, the proposed NPFC method is in a more general form since both input-like and outputlike characteristic variables are incorporated. In addition, by allowing some of the training observations to be misclassifed, the convex- and nonconvex-NPFC methods are extended from a hard nonparametric frontier to a soft one, which also provides a more conservative boundary enclosing the normal class. Both simulation studies and a real-life data set are used to evaluate and compare the proposed NPFC methods to some well-established methods in the literature. The results show that the proposed NPFC methods have competitive classifcation performance and have consistent advantages in detecting abnormal samples, especially the nonconvex-NPFC methods.Lire moins >
Langue :
Anglais
Vulgarisation :
Non
Collections :
Source :