These days, the definition of "SUV" has been getting blurrier as these vehicles become more and more popular, with some compact hatchbacks (such as the Kia Soul) being marketed as such, despite Some people saying otherwise. Thus I've decided to solve this issue with the power of machine learning, because when life gives you vague definitions, you make logistic regression.
Here is a more readable form of the dataset, which includes every car from Honda, Ford, and BMW currently being sold in the United States during the 2019 model year, with a separate entry for every for every possible drivetrain setup/bodystyle. For the purposes of this dataset, a car is an SUV if its Wikipedia page lists it as such.
As it turns out, the Kia Soul is not an SUV according to my model