Machine Learning with R Cookbook - Second Edition

How it works...

The two main constraints of association mining are to the choices regarding support and confidence. For example, if you use a high support threshold, you might remove rare item rules without considering whether these rules have a high confidence value. On the other hand, if you choose to use a low support threshold, the association mining can produce huge sets of redundant association rules, making them difficult to utilize and analyze. Therefore, we need to prune redundant rules so we can discover meaningful information from these generated rules.

In this recipe, we demonstrate how to prune redundant rules. First, we search for redundant rules. We sort the rules by a lift measure and then find subsets of the sorted rules ...