First word "genetic", genetic is the science of understanding how gene
of living things work. Genes transmit character of parents to their
offsprings. Genes compose of DNA, DNA has double helix structure.
In DNA, the four basic alleles : C,G,T,A encode the instructions to produce
all kinds of protein in living body. The real living gene is very
complex. Genes in parents can be carried to their offsprings
by various methods including : recombination and mutation. The
process of genetic in animals is very complex, many of which is still unknown
to us. Second word "algorithm" means the finite step of procedure
for a computer to carry out to solve some problem.
GA is not about genetic. The natural genetic is far more complex than what we can hope to find algorithm to imitate it. Instead, GA has been INSPIRE by natural genetic. GA uses randomization and genetic inspired method to search for the problem space. GA uses basic genetic operations : reproduction, crossover and mutation to improve the solution and use "fitness" function to guide the search.
GA can be regarded as an optimization technique. GA searches in
problem space to find the optimal solution. Optimization is a mathematical
method to solve for a solution with the constraints and to minimize or
maximize objective function. For example, solve for x, y where
x + y < 10
with objective function
min 5x + 2y