Machine Learning for Data Mining

Machine learning models

Like both statistical and decision tree models, machine learning models identify which predictors are the most important in the model.

In the following example of a neural net model, we see that the Speakers variable was the most important, followed by Premier, followed by TVs:

So, those are the most important predictors and that's all we get. We don't get any equations or a rule. We don't even know the direction of the relationship, for example, are we more likely to keep or lose customers that buy more speakers? We really don't know; we don't get that kind of information from any machine learning model because they're ...