There are two conflicting definitions. One applies a "right-hand rule": if the fingers of your right hand indicate the direction of rotation, your thumb points north. The other is the International Astronomers Union (IAU) convention: the hemisphere of a planet that is mostly above the "invariable plane" (basically, the ecliptic plane) is the "northern" hemisphere. ("Above", by yet another convention, is the side where Earth's northern hemisphere is found.) Saying that Uranus has a "97.8° tilt" puts the (right-hand rule) "north" pole on the southern side of the ecliptic, contravening the IAU convention. (Which apparently does apply to Venus.)
Joe Sixpack couldn't care less, but you'd think astronomers would have come to an agreement.
I see a reference to the "North pole of Europa" in another Space.com article, which makes me look forward to designating the poles of Uranus' moons.