Machine Learning for Predicting the Fraudiness of a Company

i was going to give a talk about predictive policing that i usually do and change the intro to be about fraud detection, because were really dealing with a lot of the same issues. And so here is my example for machine learning. Again, no equations, no greek letters, no anything. Let's say that we have investigated all these companies in the last year, and we're really predicting the fraudiness of a company. So this red line just corresponds to our prediction for any given value x. For x values between four and six, these are where we think the fraudiest companies, the company's the fraudiness, are likely to live.