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

(Brent) #1

with one of the single-attribute evaluators in the lower part of Table 10.9—not
an attribute subset evaluator.Rankernot only ranks attributes but also performs
attribute selection by removing the lower-ranking ones. You can set a cutoff
threshold below which attributes are discarded, or specify how many attributes
to retain. You can specify certain attributes that must be retained regardless of
their rank.


10.8 ATTRIBUTE SELECTION 425

Free download pdf