Genetic algorithm
It is a paradigm of machine learning that generates the behavior
patterns from the representation of evolution mechanisms. It is achieved
through generation on the inside of the machine of the population of
entities identified via genetics. People in the population will undergo
a phase of mutation. It should be noted that growth will not be an
assisted mechanism. Therefore, there is no evidence to support the
theory that the purpose of evolution is to generate humans. So, the
mechanisms of existence appear to come away to various Persons competing
for services in the World. The following steps are used to obtain
fitness using GA (Figure 4).
1. Consider populations p randomly.
2. Obtain the fitness of the population.
3. Repeat from Step 4 to 7 until convergence.
4. Choose any parent from the population individually.
5. Generates a new population through the crossover process.
6. Insert random genes in a new population to perform mutation.
7. Obtain fitness for newly generated populations.
The whole process is described as follows.
Suppose there is a target string, its goal is to produce target string
starting from a random string with the same length. Its process for
implementation is as follows.
1. The characters A-Z, a-z, 0-9, and other special symbols are
considered as genes in GA.
2. The string generated by these characters is considered a chromosome.
The fitness score is the number of characters that are differed from
characters in target string at a specific index of the string.
Therefore, individuals having lower fitness value is given more
preferences. According to the output of the algorithm, this approach
having issues in the optimal solutions so that further improvement is
needed to update the fitness score.