Pengklasifikasian Status Gizi Balita di Puskesmas Sempaja Samarinda menggunakan Probabilistic Neural Network (PNN) Tahun 2019

Abstract

Probabilistic Neural Network (PNN) is a model in Artificial Neural Networks (ANN) that is used for classification. PNN depends on the smoothing parameter (α). PNN has the advantage of being able to value of problems that previously existed in the back propagation method of ANN. The PNN method in this study was applied to the nutritional status of toddlers. Assessment of the nutritional status of toddlers can be determined through measurements of the human body known as anthropometry. Parameters for determining nutritional status based on anthropometry are age, weight and height. Therefore, in this study, a classification of the nutritional status of children under five is carried out to determine whether the toddler is experiencing good nutrition or poor nutrition. It was found that PNN with the best classification accuracy rate on the nutritional status of toddlers, namely the proportion of training data and testing data of 50%: 50% with α = 1, with accuracy results between training data and training data of 85% and accuracy results between data testing of the training data by 70%.