ContextActionService essentially puts context into UserInputService, this means it is easier to bind and unbind your actions if you use ContextActionService under conditions as you can do specific keys much easier. ContextActionService is also used for when you want actions to be enabled under certain conditions or time frames. You can read more about this in the documentation for it here.
An example is a door in a prison game, when a player is not a prisoner you might want them to have a bind to open the cell door - but when they are a prisoner we might want to remove that bind.
UserInputService is used more frequently from what I know, its a less advanced form of ContextActionService in that you’ll have events such as InputBegan and InputEnded which are useful in some cases. Whereas the action service will latch onto specific keys easily.
It all depends on the behavior you want to achieve.