Data Mining: Practical Machine Learning Tools and Techniques, Second Edition

(Brent) #1

tion theory is covered by Egan (1975); this work has been extended for visual-
izing and analyzing the behavior of diagnostic systems (Swets 1988) and is also
used in medicine (Beck and Schultz 1986). Provost and Fawcett (1997) brought
the idea of ROC analysis to the attention of the machine learning and data
mining community. Witten et al. (1999b) explain the use of recall and precision
in information retrieval systems; the F-measure is described by van Rijsbergen
(1979). Drummond and Holte (2000) introduced cost curves and investigated
their properties.
The MDL principle was formulated by Rissanen (1985). Kepler’s discovery of
his economical three laws of planetary motion, and his doubts about them, are
recounted by Koestler (1964).
Epicurus’s principle of multiple explanations is mentioned by Li and Vityani
(1992), quoting from Asmis (1984).


5.11 FURTHER READING 185

Free download pdf