PENERAPAN ALGORITMA NAIVE BAYES DAN K-NEAREST NEIGHBOR PADA KLASIFIKASI STATUS PERTUMBUHAN GIZI IBU HAMIL DI PUSKESMAS KECAMATAN CICURUG

Nutrition is a very important factor for the human body, especially for pregnant women. The health and nutrition of pregnant women has a significant impact on the development of the child, and the quality of nutrition received, because if nutritional needs are not met, the risks for the mother and baby can increase significantly. According to data from the 2018 Basic Health Survey (Riskades), 48.9% of pregnant women, 17.3% of whom suffer from chronic energy deficiency (KEK), and 28% of pregnant women are at risk of experiencing birth complications that can cause death. Even though this figure shows a decline every year, the problem of poor nutrition in pregnant women remains a major concern. In data analysis, the classification technique with the best performance was used to classify the nutritional status of pregnant women. Classification of nutritional status of pregnant women using the supervised learning method with the Naïve Bayes algorithm and K-nearest neighbor (K-NN). The data set used was 850 pregnant women in the Cicurug Community Health Center area. This data includes the variables body weight (BB), upper arm circumference (LiLA), hemoglobin (Hb), and Body Mass Index (BMI). The research results show that the Naive Bayes algorithm has an accuracy value of 79.18% with an error value of (0.6802) with the K-NN model k=3, k=5, and k=7. Meanwhile, the K-NN k=3 and k=5 algorithms have the most optimum accuracy, namely 94.92% with an error value of (0.1878), while the K-NN k=7 model has an accuracy value of 93.90% with an error value of (0.2284).
Keywords: Classification, Naive Bayes, K-NN, Nutritional Status, KEK