Adiati, Hanny (2014) Deteksi Obesitas pada Anak Usia 6-12 tahun di Indonesia Menurut Kontribusi Macronutrients (Energi, Protein, Lemak, Karbohidrat, Serat) dengan Metode Regresi Logistik Biner dan Neural Networks (SOM-RBFNN). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1310100058-Undergraduate_Thesis.pdf - Accepted Version Download (2MB) | Preview |
Abstract
Obesitas pada anak merupakan salah satu masalah
kesehatan di Indonesia. Obesitas tergolong penyakit yang
memiliki penyebab multifaktoral yang salah satu faktor
penyebabnya ditentukan oleh asupan gizi yang masuk dalam
tubuh. Macronutrients (energi, protein, lemak, karbohidrat
dan serat) yang masuk dalam tubuh dapat menjadi salah satu
faktor gizi utama untuk mendeteksi status obesitas pada anak.
Regresi Logistik biner dan Radial Basis Neural Networks
(RBFNN) merupakan metode yang sering digunakan dalam
klasifikasi dan deteksi penyakit. Salah satu yang menjadi hal
penting dalam RBFNN adalah inisialisasi pembobot awal.
Self Organizing Maps (SOM) merupakan salah satu metode
untuk mendapatkan inisialisasi pembobot awal pada RBFNN
(SOM-RBFNN). Dalam peneltian ini, didapatkan hidden layer
yang memberikan hasil terbaik adalah 2 hidden layer.
Kesimpulan akhir yang didapatkan adalah regresi logistik
biner dan neural networks dengan SOM-RBFNN memberikan
ketepatan klasifikasi yang tidak jauh berbeda dalam upaya
deteksi obesitas berdasarkan kontribusi macronutrients pada
anak usia 6-12 tahun di Indonesia.
========================================================================================================
Nowaday obesity is become a common problem in Indonesia.
Childhood obesity was become a serious problem because have a latent
effect for their future. Obesity is classified as a disease which have a
multifactoral causes that one of contributing factor is determined by the
intake of nutrients which actually consume by human. Macronutrients
(energy, protein, fat, carbohydrate and fiber) that enters the body can be
one of the major nutritional factor to detect the status of obesity in
children. Binary Logistic Regression and Radial Basis Neural Network
(RBFNN) is a method often used in classification and detection of
disease. Self Organizing Mapping (SOM) is one method to get the initial
weighting of RBFNN. The aim of this study was to classify obesity by
binary logistic regression and neural networks. All subjects are children
6-12 years old measured by ministry of health in Indonesia. This
research effort to detect obesity using binary logistic regression method
and RBFNN based SOM (SOM-RBFNN). In this research 2 hidden layer
known as the best node in hidden layer of SOM-RBFNN. The conclusion
is neural networks and logistic regression both were good classifier for
obesity detection and they were not significantly different in
classification obesity of children 6-12 years age in Indonesia based on
macronutrients contribution (Energy, Protein, Fat, Carbohydrate,
Fiber).
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.536 Adi d 3100014056693 |
| Uncontrolled Keywords: | Obesitas anak, Macronutrientst, Klasifikasi, Regresi Logistik Biner, Self Organizing mapping (SOM), Radial Basis Function Neural Network (RBFNN), Ketepatan Klasifikasi. |
| Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
| Divisions: | Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis |
| Depositing User: | Yeni Anita Gonti |
| Date Deposited: | 30 Dec 2020 03:57 |
| Last Modified: | 30 Dec 2020 03:57 |
| URI: | http://repository.its.ac.id/id/eprint/82351 |
Actions (login required)
 |
View Item |