<![CDATA[Nutritional problems in the critical period of toddlers aged 0-59 years have a fatal impact on a person's growth and development in the future. Due to lack of or excess nutrition in the first 2 years of life, it causes permanent brain function impairment and degenerative diseases. In this case, the monitoring and examination of the nutritional status of children under five carried out by health workers and posyandu cadres are generally carried out by manual recording and further analysis is carried out by comparing the nutritional measurement data of children under five with nutritional status standards. The manual analysis is prone to errors of inaccuracy in identifying the nutritional status of children under five and takes a long time due to the large amount of data, so it's not practicable. Based on these problems, the authors apply the random forest method which is optimized with genetic algorithms to classify the nutritional status of toddlers accurately and quickly. After testing, the average accuracy by the random forest method optimized with genetic algorithms is 89.58%, the average precision is 74,34%, the average recall is 58,68%, and the f1-score is 65,54% with a population size parameter of 20, 3 iterations, a crossover rate value of 0,7, an mutation rate value of 0,3, and the number of features 4. From the evaluation results of accuracy, precision, recall and f1-score obtained in this study, it shows that the genetic algorithm is able to find optimal parameters from random forests so as to produce higher accuracy, precision, recall and f1-score.]]>
Copyrights © 2022
