Clustering Numerical Data Using Artificial Ants

International audienceIn this paper we present a summary of our work which has proposed a new model for the construction of dynamics graphs. We present the state of the art in several approaches in swarm intelligence and clustering where researchers get inspiration from the numerous abilities of ants. Then we detail the construction of a chemical odor in ant species used to determine the ant nest membership in real ants from which our model is derived. We describe our main algorithm, called CL-Ant, where each ant represents one datum and the moves of ants aim at creating homogeneous groups of data that evolve together in a graph environment. This algorithm is based on the concept of artificial ants that move together in a complex manner with simple localization rules. We present an incremental extension, called CL-AntInc to treat data streams. We provide a detailed analytical comparison of our results with those obtained by other methods, using numerical datasets extracted from the Machine Learning Repository