• Early rule learning algorithms were plagued by a couple of problems.• First, they were notorious for being slow, which made them ineffective for the increasing number of large datasets.• Secondly, they were often prone to being inaccurate on noisy data.• A first step toward solving these problems is the Incremental Reduced Error Pruning (IREP) algorithm uses a combination of pre-pruning and post-pruning methods that grow very complex rules and prune them before separating the instances from the full dataset.