Pruning association rules The number of rules generated by ARM can be very large. It is necessary to prune the set of association rules by removing redundant information in order to make the classification more efficient. Given two rules R1: X1 → Y1 and R2: X2 → Y2, we define: Definition 1. The significance of a rule: R1 is more significant than R2 if and only if either (1) conf (R1) > conf (R2) or (2) conf (R1) = conf (R2) but sup(R1) > sup(R2) or (3) R1 has fewer attributes in its left hand side than R2 ◇ Definition 2. General rule: Given two rules R1: X1 → Y1 and R2: X2 → Y2, R1 is a general rule if and only if X1 ⊆ X2 ◇ Definition 3. Overlapping rule: Given two rules R1: X1 → Y1 and R2: X2 → Y2, then R3: X1 ∨ X2 → Y1(conf (R1), sup(R1)) ∨ Y2(conf (R2), sup(R2)) is an overlapping rule if and only if X1 = X2 and Y1 ≠ Y2 ◇ If the body of a rule R1 is identical to the body of a rule R2 and the head of rule R1 is inconsistent with that of rule R2, then an overlapping rule R3 between two different PPI types can be identified. Overlapping rules can be considered as common rules between two or more PPI types. On the other hand unique rules are distinctive patterns which can be used to classify interaction sites into different PPI types. We then evaluated the following condition in order to prune the set of association rules previously generated. Given two rules R1 and R2, where R1 is a general rule w.r.t. R2, ARBC eliminate R2 if R1 has more significance than R2. Sets of unique and overlapping rules were generated with the pruning procedure used in the classification.