Usage of both crossover and mutation in Genetic Algorithm
If we use both crossover and mutation in GA then this strategy always makes us closer to our desired solution because in Genetic Algorithm we start and initialize population. Then we evaluate fitness of population. Here we are using obviously characteristics from the parents but still we have not guaranteed of getting solution even we repeat this step again and again. Because as we are using crossover by joining one’s head to the tail of the other. But here we can not know that which portion is one of our desired value or which is not? On every repetition we have involved inheritance from parents but by using mutation we can avail the inheritance and randomness involvement as well which will keep changing few characteristics or bits randomly so by this we will be getting closer to the solution instead of just keeping the repetition over and over to get the success.
Last updated: March 19, 2014