Your Features Are Important? It Doesn’t Mean They Are Good | by Samuele Mazzanti | Aug, 2023

“Function Significance” just isn’t sufficient. You additionally want to take a look at “Error Contribution” if you wish to know which options are useful on your mannequin.

The idea of “characteristic significance” is extensively utilized in machine studying as essentially the most primary sort of mannequin explainability. For instance, it’s utilized in Recursive Function Elimination (RFE), to iteratively drop the least essential characteristic of the mannequin.

Nonetheless, there’s a false impression about it.

The truth that a characteristic is essential doesn’t suggest that it’s useful for the mannequin!

Certainly, after we say {that a} characteristic is essential, this merely signifies that the characteristic brings a excessive contribution to the predictions made by the mannequin. However we must always contemplate that such contribution could also be flawed.

Take a easy instance: a knowledge scientist by accident forgets the Buyer ID between its mannequin’s options. The mannequin makes use of Buyer ID as a extremely predictive characteristic. As a consequence, this characteristic could have a excessive characteristic significance even whether it is really worsening the mannequin, as a result of it can’t work nicely on unseen knowledge.

To make issues clearer, we might want to make a distinction between two ideas:

• Prediction Contribution: what a part of the predictions is as a result of characteristic; that is equal to characteristic significance.
• Error Contribution: what a part of the prediction errors is as a result of presence of the characteristic within the mannequin.

On this article, we’ll see how one can calculate these portions and how one can use them to get beneficial insights a couple of predictive mannequin (and to enhance it).

Suppose we constructed a mannequin to foretell the revenue of individuals primarily based on their job, age, and nationality. Now we use the mannequin to make predictions on three individuals.

Thus, we have now the bottom fact, the mannequin prediction, and the ensuing error: