Klasifikasi Status Gizi Balita Menggunakan Metode SVM-PSO

Eka Oktavianingsih, Khailila (2024) Klasifikasi Status Gizi Balita Menggunakan Metode SVM-PSO. Undergraduate thesis, UPN "Veteran" Jawa Timur.

	Text (COVER) 20081010146.-cover.pdf Download (782kB)
	Text (BAB 1) 20081010146.-bab1.pdf Download (19kB)
	Text (BAB 2) 20081010146.-bab2.pdf Restricted to Repository staff only until 19 September 2026. Download (355kB)
	Text (BAB 3) 20081010146.-bab3.pdf Restricted to Repository staff only until 19 September 2026. Download (452kB)
	Text (BAB 4) 20081010146.-bab4.pdf Restricted to Repository staff only until 19 September 2026. Download (783kB)
	Text (BAB 5) 20081010146.-bab5.pdf Download (24kB)
	Text (DAFTAR PUSTAKA) 20081010146.-daftarpustaka.pdf Download (132kB)

Abstract

Classification of toddler nutritional status is an important aspect in monitoring child growth and development. Problems of nutritional status that are not detected early can have an impact on children's health and development in the long term. Therefore, an accurate classification model is needed to detect nutritional status, which can assist in decision making related to nutrition. This study aims to develop a classification model of toddler nutritional status using the Support Vector Machine (SVM) method optimized with Particle Swarm Optimization (PSO). SVM was chosen because of its ability to handle non�linear data, while PSO was used to optimize SVM parameters, especially C, to improve model performance. The results showed that the application of the SVM method optimized with PSO succeeded in increasing the accuracy of toddler nutritional status classification to 92%, compared to a single SVM which only achieved an accuracy of 85%. Model evaluation using metrics showed an increase in precision and recall in the 'Normal' and 'Stunted' classes, with an increase in classification accuracy in the 'Severely Stunted' class by 10% compared to standard SVM. PSO proved effective in optimizing the C parameter in SVM. The proposed solution has proven to be effective in handling large data and can be used as a tool in clinical decision making related to the nutritional status of toddlers.

Item Type:

Thesis (Undergraduate)

Contributors: