Analysis To Predict The Quality Of Toddler Growth By Implementing The KNN And Naïve Bayes Methods

Abstract

In particular, stunting and being under the Red Line (BGM) are significant issues for society and the healthcare system. This research utilizes machine learning, particularly the K-Nearest Neighbor (KNN) and Naïve Bayes algorithms, for classifying the health of children experiencing stunting or BGM. The training data used comes from the Indonesian Posyandu website, serving as the foundation for classifying new data. This research not only identifies patterns in the data through KNN but also compares the prediction results between KNN and Naïve Bayes in assessing the probability of stunting or BGM in children. This issue reflects nutritional deficiencies and has the potential to cause developmental delays and long-term health impacts. This approach allows for the comparison of predictive outcomes, enhancing the accuracy of children's health assessments. By using the RapidMiner application, the accuracy result for KNN is 70.62% and for Naïve Bayes is 99.47%, providing a deeper understanding of the effectiveness of each algorithm in addressing child health challenges. The aim of this research is to classify new toddler data using the KNN and Naïve Bayes methods, implemented in the form of a Visual Basic application. It is hoped that this will help monitor children's health more effectively and be more easily accessible to interested parties.