A new approach to solve the traveling salesman problem by using the improved Kohonen´s self-organizing feature map

This paper proposes methods that would develop the ability of Kohonen´s self-organizing features map (SOFM) to solve optimization problems and also shows how useful SOFM is in solving optimization problems. The authors focused on the traveling salesman problem (TSP) as a typical example of an optimization problem. The conventional SOFM can solve the TSP. But the solution is not the optimum solution because the path intersects itself. Therefore, the authors propose new methods to keep the path from intersecting itself at all times. By adding these methods to the rule of changing synaptic strengths, the path length is improved by decreasing the iteration time and increasing the convergence rate