Association rule mining is an important data mining task that discovers relationships among items in a transaction database. Most approaches to association rule mining assume that the items within the dataset have a uniform distribution. Therefore, weighted association rule mining (WARM) was introduced to provide a notion of importance to individual items. In previous work most of these approaches require users to assign weights for each item. This is infeasible when we have millions of items in a dataset. In this paper we propose a novel method,
Weighted Association Rule Mining using Particle Swarm Optimization (WARM SWARM)
, which uses particle swarm optimization to assign meaningful item weights for association rule mining.