Quantum-Inspired Automatic Clustering Technique Using Ant Colony Optimization Algorithm
Quantum computing has emerged as the most challenging field of research in efficient computation. This chapter introduces a novel quantum-inspired ant colony optimization technique for automatic clustering. This chapter presents an application of this proposed technique to the automatic clustering of real-life gray-scale image data sets. In contrary to the other techniques, the proposed one requires no previous knowledge of the data to be classified. It finds the optimal number of clusters of the data by itself. The Xie-Beni cluster validity measure has been employed as the objective function for clustering purpose. Effectiveness of the proposed technique is exhibited on four real-life gray-scale images. Superiority of the proposed technique is established over its counterpart with respect to various aspects, which include accuracy, stability, computational time and standard errors. Finally, a statistical supremacy test, called unpaired two-tailed t-test, is conducted between them. It shows that superiority in favor of the proposed technique is established.