Genetic Algorithm for solving Nutrition Diet for Women Model with USDA National Nutrient Database
The objective is to find the min calories and max protein for modeling. we use NSGA-II with some preparation of data.
For methodology
[1]. uniform crossover and swap mutation as method.
[2]. tournament selection with 20% competition.
[3]. Set constrain with fat, calories, protein, carbo and fiber.
- crossover rate : 80%
- mutation rate : 20%
- generation : 100
- chromosome = 6 genes (each gene represent raw data from dataset)
- population (pool size) = 50
*NOTE : This project do speed up hyper parameter for steep over local optimal in some case.