Ant Colony Optimization for First-Order Rule Discovery

In the past, ant colony optimization has been applied to learning sets of propositional rules. In this paper, we present an algorithm for learning sets of first-order rules with ant colony optimization. First-order rules can sometimes provide a more intuitive and accurate concept description as they are more expressive than traditional propositional rules. As a case study, we apply our algorithm to expressive music performance modeling, one of the most challenging problems in music informatics, and compare our results with the results obtained by state-of-the-art first-order rule learning algorithms.