Un test statistique pour la détection d'anomalies basé sur l'erreur de reconstruction de l'ACP à noyau

Chloé Friguet 1 Laetitia Chapel 1
1 OBELIX - Environment observation with complex imagery
UBS - Université de Bretagne Sud, IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : Anomaly detection aims at declaring a query point as "normal" or not with respect to a nominal model. A non-parametric statistical test that allows the detection of anomalies given a set of (possibly high dimensional) sample points drawn from a nominal probability distribution is presented. Its test statistic is based on the distance between a query point, mapped in a feature space, and its projection on the eigen-structure of the kernel matrix computed on the sample points. The method is tested on both articial and benchmarked real data sets and demonstrates good performances regarding both type-I and type-II errors w.r.t. competing methods. This communication is based on a recently published paper by the same authors.
Liste complète des métadonnées

Cited literature [9 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01436045
Contributor : Chloé Friguet <>
Submitted on : Monday, January 16, 2017 - 10:26:31 AM
Last modification on : Thursday, February 7, 2019 - 4:53:23 PM
Document(s) archivé(s) le : Monday, April 17, 2017 - 2:20:50 PM

File

submission_170.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01436045, version 1

Citation

Chloé Friguet, Laetitia Chapel. Un test statistique pour la détection d'anomalies basé sur l'erreur de reconstruction de l'ACP à noyau. 48ème journées de Statistique, SFdS, May 2016, Montpellier, France. ⟨hal-01436045⟩

Share

Metrics

Record views

435

Files downloads

399