International Conference on Computer Technology and Development, 3rd (ICCTD 2011)
26 Study of the Effects of Variations in Crossover and Mutation Probabilities on SGA Algorithm
Download citation file:
- Ris (Zotero)
- Reference Manager
The SGA (Simple Genetic Algorithm) depends on many factors and the two most important ones are Crossover Probability and Mutation Probability. In a successful run of SGA, the average fitness of the population after mating is better than the initial average fitness i.e. fnew>finit Where fnew is average fitness after mating and finit is average fitness before mating. The average fitness is defined as the average of fitness of chromosomes in the mating pool. The chromosomes are implemented as bit strings (i.e. 1001). These bit strings are represented using integers.The performance of SGA is evaluated in this paper by varying the crossover probability and mutation probability. The results are compared with the usual SGA.