RAZAK, INNEKE FARADILA (2026) ANALISIS KINERJA ALGORITMA NAIVE BAYES DALAM KLASIFIKASI RISIKO STUNTING PADA BALITA. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
Cover.pdf Download (360kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
Bab 1.pdf Restricted to Registered users only Download (108kB) |
|
|
Text (BAB II)
Bab 2.pdf Restricted to Registered users only Download (452kB) |
|
|
Text (BAB III)
Bab 3.pdf Restricted to Registered users only Download (424kB) |
|
|
Text (BAB IV)
Bab 4.pdf Restricted to Registered users only Download (469kB) |
|
|
Text (BAB V)
Bab 5.pdf Restricted to Registered users only Download (33kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar Pustaka.pdf Restricted to Registered users only Download (128kB) |
|
|
Text (LAMPIRAN)
Lampiran.pdf Restricted to Registered users only Download (361kB) |
Abstract
Stunting is a long-term nutritional problem in toddlers that can affect physical growth, mental development, and the quality of human resources in the future. Early detection of stunting risk is an important step so that nutritional interventions can be carried out quickly and accurately, particularly during the critical period of child growth. This study aims to analyze the performance of the Naive Bayes algorithm in classifying stunting risk in toddlers based on anthropometric data and birth conditions, and to compare it with the Support Vector Machine (SVM) algorithm. The data used are secondary data from Posyandu facilities in Tangerang Regency for the 2022-2024 period, which show a fairly significant class imbalance, as only about 5.47% of toddlers in the total dataset are at risk of stunting. To address this issue, SMOTE and SMOTE-ENN techniques were applied during the data preprocessing stage. The test results show that the Naive Bayes algorithm obtained an accuracy of 76.64%, precision of 13.60%, recall of 60.78%, an F1-score of 22.22%, and a ROC-AUC of 0.7900 at the optimal threshold (0.75), following feature addition, the application of SMOTE-ENN, feature selection, and probability calibration. On the other hand, SVM showed far better performance, with an accuracy of 99.25%, precision of 94.00%, recall of 92.16%, an F1-score of 93.07%, and a ROC-AUC of 0.9834 at the optimal threshold (0.372). SVM showed superiority over Naive Bayes across all evaluation metrics, particularly in precision and its ability to distinguish between imbalanced classes. Keywords: Stunting, Naive Bayes, Support Vector Machine, SMOTE, SMOTEENN. Stunting merupakan permasalahan gizi jangka panjang pada balita yang dapat berpengaruh terhadap pertumbuhan fisik, perkembangan mental, dan kualitas sumber daya manusia di masa depan. Penemuan awal dari risiko stunting menjadi langkah penting agar intervensi gizi dapat dilakukan dengan cepat dan tepat, terutama pada masa pertumbuhan anak yang kritis. Penelitian ini bertujuan untuk menganalisis kinerja algoritma Naive Bayes dalam mengklasifikasikan risiko stunting pada balita berdasarkan data antropometri serta kondisi saat lahir, dan membandingkannya dengan algoritma Support Vector Machine (SVM). Data yang digunakan adalah data sekunder dari fasilitas Posyandu Kabupaten Tangerang pada periode 2022-2024, yang menunjukkan ketidakseimbangan kelas yang cukup signifikan karena hanya sekitar 5,47% balita yang berisiko mengalami stunting dari total data. Untuk menangani masalah ini, digunakan teknik SMOTE dan SMOTEENN dalam tahap pra-pemrosesan data. Hasil pengujian memperlihatkan bahwa algoritma Naive Bayes memperoleh accuracy 76,64%, precision 13,60%, recall 60,78%, F1-score 22,22%, dan ROC-AUC 0,7900 pada threshold optimal (0,75), setelah penambahan fitur, penerapan SMOTEENN, pemilihan fitur, dan kalibrasi probabilitas. Di sisi lain, SVM menunjukkan performa yang jauh lebih baik dengan accuracy 99,25%, precision 94,00%, recall 92,16%, F1-score 93,07%, dan ROC-AUC 0,9834 pada threshold optimal (0,372). SVM menunjukkan keunggulan dibanding Naive Bayes dalam semua metrik evaluasi, khususnya pada precision dan kemampuan untuk membedakan kelas yang tidak seimbang. Kata kunci: Stunting, Naive Bayes, Support Vector Machine, SMOTE, SMOTEENN
Actions (login required)
 |
View Item |
