Socrates is a man. All men are mortal. Therefore, Socrates is mortal.
You probably know that a general inference engine may carry out this sort of reasoning by combining the knowledge encapsulated in a rule like Human(x) -> Mortal(x) with the knowledge of the fact Human(Socrates) to infer Mortal(Socrates).
But in this exercise, rules refer only to concrete propositions given as simple unitary variables, not quantified expressions. The conjunction of all the premises appearing on the left hand side of a rule implies that the single conclusion appearing on the right hand side can be derived. We can represent rules of this form in EDN like so: [[:socrates-is-a-man :all-men-are-mortal] :-> :socrates-is-mortal].
Given this set of two rules:
[[:a :b] :-> :c]
[[:c :d :e] :-> :f