A Synergy of Artificial Bee Colony and Genetic Algorithms to Determine the Parameters of the Σ-Gram Distance

Abstract

In a previous work we presented the Σ-gram distance that computes the similarity between two sequences. This distance includes parameters that we calculated by means of an optimization process using artificial bee colony. In another work we showed how population-based bio-inspired algorithms can be sped up by applying a method that utilizes a pre-initialization stage to yield an optimal initial population. In this paper we use this pre-initialization method on the artificial bee colony algorithm to calculate the parameters of the Σ-gram distance. We show through experiments how this pre-initialization method can substantially speed up the optimization process.