A Clustering Method Based on the Artificial Bee Colony Algorithm for Gas Sensing

The authors present a method for finding the centers of clusters of complex data, where the classes are known a priori. This method uses the metaheuristic algorithm of the Artificial Bee Colony (ABC) algorithm, where the objective function is formulated according to the optimal K-means criteria. The used data come from a characterization laboratory, which deals with sensing gases using arrays of metal oxide devices. It is selected ethylene, methane and carbon monoxide for this clustering method. The main parts of the ABC algorithm are described in detail.