Chapter 10. A Different Outlook to Problems with Classification Models
Now that you have the instruments to interpret the results of data mining models, it is time to move on to executing the data modeling strategy you defined with Andy. Here, you will look at classification models, first of all understanding why they were developed and in what kinds of problems they can be useful.
You will then look at three of the most common models employed within this field, which are logistic regression, support vector machines, and random forest, carefully evaluating what the assumptions are to be satisfied in order for the model to be useful.
One note of warning before leaving you again with Andy—some of the models we are going to see here as classification models are actually sometimes employed, with slight modifications, as regression models. You should therefore not be too rigid in classifying those models into your memory. The same holds for these models being supervised, since unsupervised versions...