Classification of Big Data Stunting in North Sumatra Using Support Vector Regression Method

Stunting in children is a serious issue in society, especially in areas with high levels of malnutrition like North Sumatra. Therefore, it is important to develop an effective approach to identify the factors contributing to stunting and predict its risks in children, considering the high prevalence of stunting in this region. The high rate of stunting in North Sumatra indicates the urgency of this problem, making research on Big Data classification using Support Vector Regression (SVR) methods highly important. This study aims to offer profound understanding into factors influencing stunting in the region, thus enabling the development of more effective and targeted intervention strategies. The objective of this research is to categorize Big Data related to stunting in North Sumatra using SVR methods, taking into account factors such as wasting and malnutrition. The main focus of this research is to identify patterns related to stunting, predict the risk of stunting in children, and design more effective intervention strategies while addressing the issues of wasting and malnutrition. The research process encompasses several steps including data collection, pre-processing to handle missing values and outliers, normalization, and the application of Support Vector Regression (SVR). The final outcomes were achieved using a Voting Classifier that integrates Support Vector Classifier (SVC), Random Forest (RF), and Gradient Boosting (GB), resulting in an accuracy rate of 91.78%. This method effectively pinpoints the main factors contributing to stunting, which supports clinical decision-making and intervention strategies. The study highlights the potential of big data and machine learning in the healthcare sector, offering a model for enhancing health services and tracking children’s health conditions.