Computational Intelligence in Stochastic Solution for Toroidal N-Queen

Abstract: Evolutionary models are now robust tool in modeling complex, dynamic and non-linear data processing due to its flexible mathematical structure to yield optimal results even with imprecise, ambiguity and noise at its input. The study investigates evolutionary models for solving Toroidal N-Queen task. Evolutionary hybrids have become veritable models for multipoint search in CSPs with application areas like image/video analysis, military, network construction, OS resource scheduling and allocation, medicine, cloud and clustering computing etc. Solution space representation and fitness functions (as common to all algorithms) are discussed. For the adopted support/confidence model, w1=0.2 and w2=0.8 respectively yields better convergence. Other suggested value combinations led to a slower or non-convergence. For 20-queen, HPSOGA and CGA found an optimal solution in 32 seconds after 188 iterations; GSAGA found its optimal solution in 18seconds after 402 iterations and consequently, GASA found an optimal solution 2.112seconds after 391 iterations respectively.

Keywords: Swarm, Agents, Particles, Elitist, Constraints, Fitness Function
¹Department of Mathematics/Computer Science, Federal University of Petroleum Resources Effurun, Delta State, ojugo_arnold@yahoo.com, arnoldojugo@yahoo.com
²Department of Computer Science, Delta State Polytechnic

[PDF]

Cite: A.A. Ojugo and R.E Yoro, "Computational Intelligence in Stochastic Solution for Toroidal N-Queen," Progress in Intelligent Computing and Applications , vol. 2, no. 1, pp. 46-56, March 2013.