SOM-based novelty detection using novel data

Authors:
Hyoung-joo Lee;Sungzoon Cho
Affiliations:
Dept. of Industrial Engineering, Seoul National University, Seoul, Korea;Dept. of Industrial Engineering, Seoul National University, Seoul, Korea
Venue:
IDEAL'05 Proceedings of the 6th international conference on Intelligent Data Engineering and Automated Learning
Year:
2005

Citing 10
Cited 8

Autoassociator-based models for speaker verification

Pattern Recognition Letters
Supervised versus unsupervised binary-learning by feedforward neural networks

Machine Learning
Soft Margins for AdaBoost

Machine Learning
Self-Organizing Maps

Self-Organizing Maps
A Mixture Approach to Novelty Detection Using Training Data with Outliers

EMCL '01 Proceedings of the 12th European Conference on Machine Learning
Novelty detection: a review—part 1: statistical approaches

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

Signal Processing
Support Vector Data Description

Machine Learning
Estimating the Support of a High-Dimensional Distribution

Neural Computation
A neural network-based model for paper currency recognition and verification

IEEE Transactions on Neural Networks

Focusing on non-respondents: Response modeling with novelty detectors

Expert Systems with Applications: An International Journal
Anomaly detection in mobile communication networks using the self-organizing map

Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology - VIII Brazilian Symposium on Neural Networks
Time Series Clustering for Anomaly Detection Using Competitive Neural Networks

WSOM '09 Proceedings of the 7th International Workshop on Advances in Self-Organizing Maps
Analysis of Time Series Novelty Detection Strategies for Synthetic and Real Data

Neural Processing Letters
Combining SOM and local minimum enclosing spheres for novelty detection

CCDC'09 Proceedings of the 21st annual international conference on Chinese control and decision conference
The novelty detection approach for different degrees of class imbalance

ICONIP'06 Proceedings of the 13th international conference on Neural Information Processing - Volume Part II
Retraining a novelty detector with impostor patterns for keystroke dynamics-based authentication

ICB'06 Proceedings of the 2006 international conference on Advances in Biometrics
Improving authentication accuracy of unfamiliar passwords with pauses and cues for keystroke dynamics-based authentication

WISI'06 Proceedings of the 2006 international conference on Intelligence and Security Informatics

Quantified Score

Hi-index	0.00

Visualization

Abstract

Novelty detection involves identifying novel patterns. They are not usually available during training. Even if they are, the data quantity imbalance leads to a low classification accuracy when a supervised learning scheme is employed. Thus, an unsupervised learning scheme is often employed ignoring those few novel patterns. In this paper, we propose two ways to make use of the few available novel patterns. First, a scheme to determine local thresholds for the Self Organizing Map boundary is proposed. Second, a modification of the Learning Vector Quantization learning rule is proposed so that allows one to keep codebook vectors as far from novel patterns as possible. Experimental results are quite promising.