Novelty detection using a new group outlier factor

Authors:
Amine Chaibi;Mustapha Lebbah;Hanane Azzag
Affiliations:
Sorbonne Paris City - CNRS, LIPN-UMR 7030, University of Paris 13, Villetaneuse, France;Sorbonne Paris City - CNRS, LIPN-UMR 7030, University of Paris 13, Villetaneuse, France;Sorbonne Paris City - CNRS, LIPN-UMR 7030, University of Paris 13, Villetaneuse, France
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part III
Year:
2012

Citing 16
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Network constraints and multi-objective optimization for one-class classification

Neural Networks
LOF: identifying density-based local outliers

SIGMOD '00 Proceedings of the 2000 ACM SIGMOD international conference on Management of data
Self-Organizing Maps

Self-Organizing Maps
The Error-Reject Tradeoff

Open Systems & Information Dynamics
Parzen-Window Network Intrusion Detectors

ICPR '02 Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02) Volume 4 - Volume 4
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Novelty detection: a review—part 1: statistical approaches

Signal Processing
Novelty detection: a review—part 2: neural network based approaches

Signal Processing
Estimating the Support of a High-Dimensional Distribution

Neural Computation
An introduction to ROC analysis

Pattern Recognition Letters - Special issue: ROC analysis in pattern recognition
Kernel PCA for novelty detection

Pattern Recognition
Robust partitional clustering by outlier and density insensitive seeding

Pattern Recognition Letters
SOMSO: a self-organizing map approach for spatial outlier detection with multiple attributes

IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Local Outlier Detection Based on Kernel Regression

ICPR '10 Proceedings of the 2010 20th International Conference on Pattern Recognition
IterativeSOMSO: an iterative self-organizing map for spatial outlier detection

ISNN'10 Proceedings of the 7th international conference on Advances in Neural Networks - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present in this paper a new measure named GOF (Group Outlier Factor) for cluster outliers and novelty detection. The main difference between GOF and existing methods is that being an outlier is not associated to a single pattern but to a cluster. GOF is based on relative density of each group of data and provides a quantitative indicator of outlier-ness which enables to detect automatically "cluster outliers". To learn GOF measure, we integrate it in a clustering process using Self-organizing Map. Experimental results and comparison studies show that the use of GOF sensibly improves the results in term of cluster-outlier detection and novelty detection.