Creating and evaluating regression algorithms
We talked about a few different algorithms in previous chapters. Which one should we choose? There are pros and cons for each, and sometimes it's not apparent which one we should go for. In this section, we'll look at a few possible algorithms and do a quick check to determine how viable each of them is. We'll then train the winner and finally analyze in more depth the results by looking at a few evaluation techniques. Before we do that, let's make sure we are looking at the correct problem family.
Comparing regression and classification
When we've looked at the target data and what our goal is, we saw that the quality is measured by discrete values from 1 to 9. If that's the case, then why aren't we looking at this as a classification problem? The short answer is we could. This example was chosen to make you think about the nuances that can arise with data science, and the answer you get depends...
