The definition of fitting is that it is finding the correct pattern in the data and when it comes to machine learning and data science that's really one of the top goals. The entire goal of what we're trying to do is we're trying to use historical data in order to drive our decisions to generate probabilities to make predictions and so fittin is the way that we can do that properly where we find the correct pattern.

If you remember back to when we walked through the support vector machine we walk through a few different examples. Each one of these lines was an attempt to fit the data to find that correct pattern. In the very first example when we tried to find the hyperplane where we tried to find the boundary between the two classes this was a poor fit because it didn't properly make the right predictions on the elements.

Then we made another attempt at finding the hyperplane this was another type of fit. This also wasn't ideal.

And then in the third example we were able to find that correct hyper plane we were able to find the boundary. And here we found the correct pattern in the data which means that we performed a fit or a fitting process that was effective.