Regularised feed forward neural networks for streamed data classification problems

Description

Streamed data classification problems (SDCPs) require classifiers with the ability to learn and to adjust to the underlying relationships in data streams, in real-time. This requirement poses a challenge to classifiers, because the learning task is no longer just to find the optimal decision boundaries, but also to track changes in the decision boundaries as new training data is received. The challenge is due to concept drift, i.e. the changing of decision boundaries over time. Changes include disappearing, appearing, or shifting decision boundaries. This thesis proposes an online learning approach for feed forward neural networks (FFNNs) that meets the requirements of SDCPs. The approach uses regularisation to optimise the architecture via the weights, and quantum particle swarm optimisation (QPSO) to dynamically adjust the weights. The learning approach is applied to a FFNN, which uses rectified linear activation functions, to form a novel SDCP classifier. The classifier is empirically investigated on several SDCPs. Both weight decay (WD) and weight elimination (WE) are investigated as regularisers. Empirical results show that using QPSO with no regularisation, causes the classifier to completely saturate. However, using QPSO with regularisation enables the classifier to dynamically adapt both its implicit architecture and weights as decision boundaries change. Furthermore, the results favour WE over WD as a regulariser for QPSO. / Dissertation (MSc)--University of Pretoria, 2020. / National Research Foundation (NRF) / Computer Science / MSc / Unrestricted

Links & Downloads

1. http://hdl.handle.net/2263/75804
2. Ellis, M 2020, Regularised feed forward neural networks for streamed data classification problems, MSc Dissertation, University of Pretoria, Pretoria, viewed yymmdd <http://hdl.handle.net/2263/75804>
3. S2020

Tags

Computational Intelligence

Data Streams

Feed Forward Neural Networks

Quantum Particle Swarm Optimisation

Regularisation

Feed Forward Neural Networks

Classification Problems

Concept drift

UCTD

Additional Fields

Identifer	oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:up/oai:repository.up.ac.za:2263/75804
Date	January 2020
Creators	Ellis, Mathys
Contributors	Bosman, Anna S., [e-mail], Engelbrecht, Andries P.
Publisher	University of Pretoria
Source Sets	South African National ETD Portal
Language	English
Detected Language	English
Type	Dissertation
Rights	© 2019 University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria.