Optimization of Support Vector Machine Algorithm Using Stunting Data Classification

Several studies from Indonesia reveal that malnutrition and stunting are still severe concerns to be addressed in the future. The complexity of problem or nutritional status requires responsibility all parties, including science technology. issue monitoring data collection related children Indonesia, especially Medan City, North Sumatra Province, is an essential factor determining calculations carried out by each Community Health Center with many attributes. Currently, Support Vector Machine method a solution increase government intervention's effectiveness classifying stunting. However, algorithm needs improve, namely difficulty selecting right optimal features for attribute weights, causing low prediction accuracy. Therefore, researchers aim optimize Algorithm Particle Swarm Optimization using Linear, Polynomial, Sigmoid, Radial Basic Function kernels. results were obtained research utilizing data, performance improving based on four kernel tests, different results, not kernels this study can improve accuracy well. best Accuracy value 78%, Precision 89%, Recall 66%, F1-Score 72%, so it feasible accurate information regarding classification status.