A Small Sphere and Large Margin Approach for Novelty Detection Using Training Data with Outliers

Authors:
Mingrui Wu;Jieping Ye
Affiliations:
Yahoo! Inc., Sunnyvale;Arizona State University, Tempe
Venue:
IEEE Transactions on Pattern Analysis and Machine Intelligence
Year:
2009

Citing 0
Cited 18

A robust fuzzy rough set model based on minimum enclosing ball

RSKT'10 Proceedings of the 5th international conference on Rough set and knowledge technology
A theoretical framework for multi-sphere support vector data description

ICONIP'10 Proceedings of the 17th international conference on Neural information processing: models and applications - Volume Part II
Multiple distribution data description learning algorithm for novelty detection

PAKDD'11 Proceedings of the 15th Pacific-Asia conference on Advances in knowledge discovery and data mining - Volume Part II
The Maximum Vector-Angular Margin Classifier and its fast training on large datasets using a core vector machine

Neural Networks
A novel parameter refinement approach to one class support vector machine

ICONIP'11 Proceedings of the 18th international conference on Neural Information Processing - Volume Part II
Generalised support vector machine for brain-computer interface

ICONIP'11 Proceedings of the 18th international conference on Neural Information Processing - Volume Part I
Peer-to-Peer network classification using nu-maximal margin spherical structured multiclass support vector machine

ICDEM'10 Proceedings of the Second international conference on Data Engineering and Management
Selective ensemble of support vector data descriptions for novelty detection

ISNN'12 Proceedings of the 9th international conference on Advances in Neural Networks - Volume Part I
Soft Minimum-Enclosing-Ball Based Robust Fuzzy Rough Sets

Fundamenta Informaticae - Rough Sets and Knowledge Technology (RSKT 2010)
Local anomaly descriptor: a robust unsupervised algorithm for anomaly detection based on diffusion space

Proceedings of the 21st ACM international conference on Information and knowledge management
Scaling up minimum enclosing ball with total soft margin for training on large datasets

Neural Networks
One-Class classification through optimized feature boundaries detection and prototype reduction

ICANN'12 Proceedings of the 22nd international conference on Artificial Neural Networks and Machine Learning - Volume Part I
Polichotomies on imbalanced domains by one-per-class compensated reconstruction rule

SSPR'12/SPR'12 Proceedings of the 2012 Joint IAPR international conference on Structural, Syntactic, and Statistical Pattern Recognition
Run-time validation of knowledge-based systems

Proceedings of the seventh international conference on Knowledge capture
Multi-local model image set matching based on domain description

Pattern Recognition
Heartbeat classification using disease-specific feature selection

Computers in Biology and Medicine
Review: A review of novelty detection

Signal Processing
Regularized correntropy criterion based feature extraction for novelty detection

Neurocomputing

Quantified Score

Hi-index	0.14

Visualization

Abstract

We present a small sphere and large margin approach for novelty detection problems, where the majority of training data are normal examples. In addition, the training data also contain a small number of abnormal examples or outliers. The basic idea is to construct a hypersphere that contains most of the normal examples, such that the volume of this sphere is as small as possible, while at the same time the margin between the surface of this sphere and the outlier training data is as large as possible. This can result in a closed and tight boundary around the normal data. To build such a sphere, we only need to solve a convex optimization problem that can be efficiently solved with the existing software packages for training \nu\hbox{-}SupportVectorMachines. Experimental results are provided to validate the effectiveness of the proposed algorithm.