I find it fascinating with these kinds of datasets that the outliers are more likely than not errors. For example, I'm guessing that "Nevskiy Prospekt" is most

I find it fascinating with these kinds of datasets that the outliers are more likely than not errors. For example, I'm guessing that "Nevskiy Prospekt" is most likely in the non-Florida St. Petersburg.