Abstract
The need to deduce interesting and valuable information from large, complex, information-rich data sets is common to many research fields. Rule discovery or rule mining uses a set of IF-THEN rules to classify a class or category in a comprehensible way. Besides the classical approaches, many rule mining approaches use biologically-inspired algorithms such as evolutionary algorithms and swarm intelligence approaches. In this paper, a Particle Swarm Optimization based discrete classification implementation with a local search strategy (DPSO-LS) was devised and applied to discrete data. In addition, a fuzzy DPSO-LS (FDPSO-LS) classifier is proposed for both discrete and continuous data in order to manage imprecision and uncertainty. Experimental results reveal that DPSO-LS and FDPSO-LS outperform other classification methods in most cases based on rule size, True Positive Rate (TPR), False Positive Rate (FPR), and precision, showing slightly improved results for FDPSO-LS.