Non-Standard Parameter Adaptation for Exploratory Data Analysis

  • Wesam Ashour Barbakh
  • Ying Wu
  • Colin Fyfe

Part of the Studies in Computational Intelligence book series (SCI, volume 249)

Table of contents

  1. Front Matter
  2. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 1-6
  3. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 7-28
  4. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 29-48
  5. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 49-72
  6. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 73-84
  7. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 85-108
  8. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 109-122
  9. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 123-149
  10. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 151-174
  11. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 175-197
  12. Wesam Ashour Barbakh, Ying Wu, Colin Fyfe
    Pages 199-205
  13. Back Matter

About this book

Introduction

Exploratory data analysis, also known as data mining or knowledge discovery from databases, is typically based on the optimisation of a specific function of a dataset. Such optimisation is often performed with gradient descent or variations thereof. In this book, we first lay the groundwork by reviewing some standard clustering algorithms and projection algorithms before presenting various non-standard criteria for clustering. The family of algorithms developed are shown to perform better than the standard clustering algorithms on a variety of datasets.

We then consider extensions of the basic mappings which maintain some topology of the original data space. Finally we show how reinforcement learning can be used as a clustering mechanism before turning to projection methods.

We show that several varieties of reinforcement learning may also be used to define optimal projections for example for principal component analysis, exploratory projection pursuit and canonical correlation analysis. The new method of cross entropy adaptation is then introduced and used as a means of optimising projections. Finally an artificial immune system is used to create optimal projections and combinations of these three methods are shown to outperform the individual methods of optimisation.

Keywords

Clustering data analysis data mining knowledge discovery machine learning principal component analysis reinforcement learning

Authors and affiliations

  • Wesam Ashour Barbakh
    • 1
  • Ying Wu
    • 2
  • Colin Fyfe
    • 3
  1. 1.Computer Engineering Department, Faculty of EngineeringIslamic University of GazaGazaPalestine
  2. 2.Coastal and Marine Resources CentreUniversity College Cork, Irish Naval Base, Haulbowline, CobhCo. CorkIreland
  3. 3.School of Computing, Applied Computational Intelligence, Research UnitUniversity of the West of ScotlandPaisleyScotland, UK

Bibliographic information

  • DOI https://doi.org/10.1007/978-3-642-04005-4
  • Copyright Information Springer Berlin Heidelberg 2009
  • Publisher Name Springer, Berlin, Heidelberg
  • eBook Packages Engineering
  • Print ISBN 978-3-642-04004-7
  • Online ISBN 978-3-642-04005-4
  • Series Print ISSN 1860-949X
  • Series Online ISSN 1860-9503
  • About this book