Knowledge Discovery and Data Mining
Presents many fundamental technologies and recent advances in data mining and knowledge discovery.
TABLE OF CONTENTS Knowledge Discovery and Data Mining
Chapter 9: Discovering Knowledge from Low-quality Meteorological Databases
C. M. Howard and V. J. Rayward-Smith
University of East Anglia, UK
9.1 Introduction
Knowledge discovery in databases (KDD) is a multistage process that, given data describing a number of past experiences of a situation, can be used to find useful knowledge in the form of patterns which may be hidden therein; this knowledge may be used to make future predictions. There is strong commercial interest in data mining because of the potential for companies to gain an advantage over their competitors. The seven key stages of KDD [1] are:
-
define problem and goals;
-
data collection and warehousing;
-
data preprocessing;
-
data mining;
-
rule analysis;
-
trial;
-
implementation.
A more detailed description of the knowledge-discovery process and, in particular, of data mining can be found in Fayyad et al. [2] and Holsheimer and Siebes [3].
9.1.1 The Meteorological Domain
Meteorological societies and universities worldwide frequently collect vast amounts of data from satellites and weather stations. Given a collection of datasets, we were asked to examine a sample of such data and look for patterns which may exist between certain geographical locations over time. Similar work has been carried out for some time using standard statistical techniques [4] and occasionally neural networks [5]; part of the aim of the work was to determine whether an approach to data mining based on rule induction using simulated annealing could be used to accomplish the same task. The datasets used throughout the work include sea and land surface temperatures [6], sea-level pressures, geomagnetic data and global...
Copyright The Institution of Electrical Engineers 1999 under license agreement with Books24x7
