Unsupervised learning with normalised data and non-Euclidean norms

Authors:
K. A. J. Doherty;R. G. Adams;N. Davey
Affiliations:
University of Hertfordshire, Hatfield, Hertfordshire, AL10 9AB, UK;University of Hertfordshire, Hatfield, Hertfordshire, AL10 9AB, UK;University of Hertfordshire, Hatfield, Hertfordshire, AL10 9AB, UK
Venue:
Applied Soft Computing
Year:
2007

Citing 6
Cited 2

Introduction to the theory of neural computation

Introduction to the theory of neural computation
Topology representing networks

Neural Networks
Clustering Algorithms

Clustering Algorithms
Self-Organizing Maps

Self-Organizing Maps
Comparing neural networks: a benchmark on growing neural gas, growing cell structures, and fuzzy ARTMAP

IEEE Transactions on Neural Networks
`Neural-gas' network for vector quantization and its application to time-series prediction

IEEE Transactions on Neural Networks

Supervised learning vector quantization for projecting missing weights of hierarchical neural networks

WSEAS Transactions on Information Science and Applications
Dimensionality Reduction with Unsupervised Feature Selection and Applying Non-Euclidean Norms for Classification Accuracy

International Journal of Data Warehousing and Mining

Quantified Score

Hi-index	0.00

Visualization

Abstract

The measurement of distance is one of the key steps in the unsupervised learning process, as it is through these distance measurements that patterns and correlations are discovered. We examined the characteristics of both non-Euclidean norms and data normalisation within the unsupervised learning environment. We empirically assessed the performance of the K-means, neural gas, growing neural gas and self-organising map algorithms with a range of real-world data sets and concluded that data normalisation is both beneficial in learning class structure and in reducing the unpredictable influence of the norm.