Combining multiple classifiers to quantitatively rank the impact of abnormalities in flight data

Authors:
E. Smart;D. Brown;J. Denman
Affiliations:
Institute of Industrial Research, Floor 7, Mercantile House, Hampshire Terrace, University of Portsmouth, Portsmouth, Hampshire, PO1 2EG, United Kingdom;Institute of Industrial Research, Floor 7, Mercantile House, Hampshire Terrace, University of Portsmouth, Portsmouth, Hampshire, PO1 2EG, United Kingdom;Flight Data Services (FDS) Ltd., 189-199 West Street, Fareham, Hampshire, PO16 0EN, United Kingdom
Venue:
Applied Soft Computing
Year:
2012

Citing 13
Cited 1

A training algorithm for optimal margin classifiers

COLT '92 Proceedings of the fifth annual workshop on Computational learning theory
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
Using an Ensemble of One-Class SVM Classifiers to Harden Payload-based Anomaly Detection Systems

ICDM '06 Proceedings of the Sixth International Conference on Data Mining
Evolving an artificial neural network classifier for condition monitoring of rotating mechanical systems

Applied Soft Computing
One-Class Novelty Detection for Seizure Analysis from Intracranial EEG

The Journal of Machine Learning Research
Artificial neural network approach for fault detection in rotary system

Applied Soft Computing
Feature-based classifier ensembles for diagnosing multiple faults in rotating machinery

Applied Soft Computing
Fault diagnosis of spur bevel gear box using artificial neural network (ANN), and proximal support vector machine (PSVM)

Applied Soft Computing
Combined use of principal component analysis and self organisation map for condition monitoring in pickling process

Applied Soft Computing
Support vector machine in novelty detection for multi-channel combustion data

ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
Structural risk minimization over data-dependent hierarchies

IEEE Transactions on Information Theory

Combining features of negative correlation learning with mixture of experts in proposed ensemble methods

Applied Soft Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel two phase method that combines one class support vector machine classifiers using combination rules to quantitatively assess the degree of abnormality at various heights during individual aircraft descents and also over the whole descent. Whilst classifiers have been combined before in the literature with success, it is the first time they have been applied to the problem of analysing the act of descending of commercial jet aircraft. The method is tested on artificial Gaussian data and flight data from an industrial partner, Flight Data Services Ltd., the world's leading flight data analysis provider, with promising results.