CD Skripsi
DIAGNOSA STUNTING PADA BALITA MENNGGUNAKAN METODE NAÏVE BAYES UNTUK SISTEM PAKAR
Stunting is a chronic nutritional problem in toddlers that negatively impacts their
physical and cognitive development, thereby affecting the quality of a nation's
human resources. At Puskesmas Pembantu Alam Raya, the previous system relied
on manual calculations using Z-Score, which could be time-consuming and often
resulted in incomplete reports submitted to the main health center, delaying
interventions for toddlers at risk of stunting. Hence, there is a critical need to
develop a more efficient and effective system to manage extensive data and
expedite the assessment of stunting nutritional status in toddlers. This research
employs the Naïve Bayes method to develop an expert system for diagnosing
stunting in toddlers, chosen for its ability to handle complex data and provide
probabilistic interpretations aiding prediction confidence. The study began with
field surveys to understand stunting issues and collect anthropometric data such as
gender, age, weight, height, head circumference, and upper arm circumference. The
data were divided into training and testing sets, and the system was modeled using
PHP programming language to integrate the Naïve Bayes method into the expert
system. Performance evaluation using a confusion matrix showed an accuracy of
97.05%, with 50% precision and 50% recall, and an error rate of 2.94%. Despite
the high accuracy, improvements in precision and recall are necessary to enhance
diagnostic reliability. This expert system is recommended for monitoring stunting
nutritional status in toddlers at Puskesmas Pembantu Alam Raya, aiming to
contribute effectively and efficiently to stunting prevention programs.
Keywords: Naïve Bayes, Stunting, Expert System, Diagnosis, Toddlers
Tidak tersedia versi lain