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

with weights w.Figure 4.9(b) shows an example of this function in one dimen-

sion, with two weights w 0 =0.5 and w 1 =1.

Just as in linear regression, weights must be found that fit the training data

well. Linear regression measures the goodness of fit using the squared error. In

logistic regression the log-likelihoodof the model is used instead. This is given

by

122 CHAPTER 4| ALGORITHMS: THE BASIC METHODS

–5

–4

–3

–2

–1

0

1

2

3

4

5

0 0.2 0.4 0.6 0.8 1

(a)

0

0.2

0.4

0.6

0.8

1

• 10 – 5 0 5 10
  (b)
  Figure 4.9Logistic regression: (a) the logit transform and (b) an example logistic regres-
  sion function.