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

the original dataset based on dummy class values ziand weights wi.We assume that p(1 |a) is computed using the fjthat were built ...

7.5 COMBINING MULTIPLE MODELS 329 overall classification. This can be done simply by voting, taking the majority vote at an opti ...

A simple example tree for the weather data is shown in Figure 7.11, where a positive value corresponds to class play =noand a ne ...

7.5 COMBINING MULTIPLE MODELS 331 every possible location in the tree is considered for addition, and a node is added according ...

Stacking Stacked generalization,or stackingfor short, is a different way of combining mul- tiple models. Although developed some ...

7.5 COMBINING MULTIPLE MODELS 333 because it will inevitably learn to prefer classifiers that overfit the training data over one ...

for this purpose. In the words of David Wolpert, the inventor of stacking, it is reasonable that “relatively global, smooth” lev ...

7.5 COMBINING MULTIPLE MODELS 335 However, we do not have to use the particular code words shown. Indeed, there is no reason why ...

We have identified one property of a good error-correcting code: the code words must be well separated in terms of their Hamming ...

7.6 USING UNLABELED DATA 337 7.6 Using unlabeled data When introducing the machine learning process in Chapter 2 we drew a sharp ...

and cluster labels are gleaned from the labeled data. The EM procedure guar- antees to find model parameters that have equal or ...

7.6 USING UNLABELED DATA 339 automatically labeled data. The solution is to introduce a weighting parameter that reduces the con ...

There is some experimental evidence, using Naïve Bayes throughout as the learner, that this bootstrapping procedure outperforms ...

deal with weighted instances in Section 6.5 under Locally weighted linear regres- sion(page 252). One way of obtaining probabili ...

Dougherty et al. (1995) give a brief account of supervised and unsupervised discretization, along with experimental results comp ...

Domingos (1997) describes how to derive a single interpretable model from an ensemble using artificial training examples. Bayesi ...

...

Machine learning is a burgeoning new technology for mining knowledge from data, a technology that a lot of people are beginning ...

learning experts, nor from the data itself, but from the people who work with the data and the problems from which it arises. Th ...

both the number of instances and the number of attributes. For top-down deci- sion tree inducers, we saw in Section 6.1 (pages 1 ...