Do you have a favorite machine learning technique?

(inspiring uplifting music) - [Man] Do you have a favorite technique? - Here is the thing what I learned very early on, is it's not about necessarily the technique, it's actually about the data coming in. Much of the lift that you get, the benefits of an algorithm that you see are actually delivered from the quality of the data. Garbage in, garbage out. And so how do you make sure that you've got a good dataset, a clean dataset? And if you can do that and you have bare minimum good quality algorithms, you're going to get a lot of lift. And then the other part of that is the feedback cycle. Once the machine learning algorithm has happened, what comes out of that machine learning or new datasets that come in, how do you pipe that and instrument that all together to stay good and stay on top of an iterative environment where things may be evolving or changing? (inspiring uplifting music)