Genetic algorithms on documents clustering

This paper deals with the documents clustering problem via evolutionary genetic algorithms. A genetic encoding method, a selection operator, a crossover operator, and a mutation operator are established and a fitness function is determined in this paper. Being different from the previous method, evolution computing is carried out before computing the fitness function. This algorithm can be extended to general combination optimization problems; especially, solution space of problems is a subset tree. Finally, a numerical example is used to illustrate the effectiveness of the methods proposed.