How it works...
When using standard techniques, we usually need to rely on very simple interactions (usually two-way interactions) that are built by adding term to the model, such as: var1 x var2. This is not very practical in many cases, when we have lots of features. The underlying algorithm for logic regression explores (as much as possible) the combination space by intelligently evaluating some of these combinations.
Unfortunately, this model reflects (again) how difficult it is to predict sports results. At the end, each game is not only determined by variables that we can measure, but also on specific game situations that are difficult to predict.
In essence, the logic regression model works by finding a set of logic expressions (
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access