Choosing the right skincare product is an important step in facial skin care, but this process requires the expertise of a specialist, which is relatively slow. The purpose of this research is to optimize the skincare product selection process by creating an expert system using the Matlab application. The method used in making the expert system is the Tsukamoto fuzzy method and genetic algorithms. The Tsukamoto fuzzy inference system helps select skincare by using membership limits according to specialist at clinic. There are four input variables, namely skin moisture, enlarged pores, blackheads, telangiactasis and one output, namely type of skincare package. The Tsukamoto fuzzy membership limit will be optimized with genetic algorithm using different crossover rate (cr) variations. The results showed that Tsukamoto's fuzzy accuracy level was 65.37%. After optimizing the membership limit, the accuracy rate increases by 17.63% to 82,93%. The increase in accuracy shows that the expert system using the two Tsukamoto fuzzy methods and the genetic algorithm is better when compared to using only the Tsukamoto fuzzy method. The optimal parameter in the genetic algorithm is the Crossover rate (cr) 0.4, with a Mutation Rate of 0.01 and the number of generations is 10.
Copyrights © 2023