Non-Standard Parameter Adaptation for Exploratory Data Analysis

Wesam Ashour Barbakh, Ying Wu, Colin Fyfe (Autoren)

Buch | Softcover

XI, 223 Seiten

2012 | 2009
Springer Berlin (Verlag)
978-3-642-26055-1 (ISBN)

Artikel merken

A review of standard algorithms provides the basis for more complex data mining techniques in this overview of exploratory data analysis. Recent reinforcement learning research is presented, as well as novel methods of parameter adaptation in machine learning.

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.

Introduction.- Review of Clustering Algorithms.- Review of Linear Projection Methods.- Non-standard Clustering Criteria.- Topographic Mappings and Kernel Clustering.- Online Clustering Algorithms and Reinforcement learning.- Connectivity Graphs and Clustering with Similarity Functions.- Reinforcement Learning of Projections.- Cross Entropy Methods.- Conclusions.

Erscheint lt. Verlag	14.3.2012
Reihe/Serie	Studies in Computational Intelligence
Zusatzinfo	XI, 223 p.
Verlagsort	Berlin
Sprache	englisch
Maße	155 x 235 mm
Gewicht	369 g
Themenwelt	Informatik ► Datenbanken ► Data Warehouse / Data Mining
	Informatik ► Theorie / Studium ► Künstliche Intelligenz / Robotik
	Mathematik / Informatik ► Mathematik ► Angewandte Mathematik
	Technik
Schlagworte	artificial immune systems • Clustering • Computational Intelligence • cross entropy • Data Analysis • Data Mining • Knowledge Discovery • machine learning • Non-Standard Exploratory Data Analysis • Principal Component Analysis • Reinforcement Learning
ISBN-10	3-642-26055-1 / 3642260551
ISBN-13	978-3-642-26055-1 / 9783642260551
Zustand	Neuware