CLASSIFICATION OF THE NUTRITION STATUS TODDLER USING THE SVM METHOD (CASE STUDY: BANJARAGUNG VILLAGE, BARENG, JOMBANG)

Abstract

Improving the health status of children under five is very necessary in determining the next generation of the Indonesian nation. One of the efforts that can be realized is to maintain the nutrition of children under five in the community. Balanced nutrition can increase immunity and increase intelligence so as to make normal growth. In social life, nutritional status is obtained through anthropometric measurements at a posyandu where people generally use the BB/U index or body weight compared to age to determine the nutritional status of toddlers. This study aims to make it easier to identify the nutritional status of toddlers using Data Mining with Support Vector Machine (SVM). system built with PHP programming language and postgreSQL database. This study uses data on 314 toddlers in 4 groups of posyandu in the village. The data was tested 2 times, the first with a 50:50 comparison and the second 70:30 for training data and testing data. The results showed an accuracy of 96% and 98%, in other words, SVM was categorized as good for testing the nutritional status of children under five.