BayiSehatKita: Aplikasi Berbasis Web Untuk Prediksi Risiko Stunting Pada Anak Usia Dini

Pujiantoro, Abraham Christopher (2021) BayiSehatKita: Aplikasi Berbasis Web Untuk Prediksi Risiko Stunting Pada Anak Usia Dini. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
05211740000049-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (2MB) | Request a copy

Abstract

Stunting (kerdil) adalah kondisi di mana balita memiliki panjang atau tinggi badan yang kurang jika dibandingkan dengan umur. Stunting memiliki dampak negatif kesehatan individu yang menderitanya dalam jangka pendek maupun panjang. Prevalensi stunting balita nasional di Indonesia menurut Riset Kesehatan Dasar (Riskesdas) 2018 adalah sebesar 30.8%. Angka ini lebih tinggi dibandingkan prevalensi stunting dunia tahun 2019 menurut World Health Organization (WHO) sebesar 21.3% dan Asia Tenggara sebesar 24.7%. Oleh karenanya diperlukan upaya untuk menurunkan angka stunting di Indonesia. Penelitian terkait stunting di Indonesia saat ini kebanyakan masih berupa analisis faktor determinan yang memengaruhi stunting dan masih sedikit penelitian yang membahas model prediksi dini risiko stunting dan mengimplementasikannya dalam bentuk aplikasi yang mudah diakses masyarakat. Tugas akhir ini bertujuan untuk mengembangkan aplikasi berbasis web untuk memprediksi probabilitas risiko stunting pada anak usia dini berdasarkan faktor determinan. Model prediksi probabilitas tersebut dibuat dengan menggunakan algoritma Naive Bayes. Tugas akhir ini juga bertujuan untuk mengekstraksi dan menyebarluaskan informasi pengaruh setiap faktor determinan terhadap probabilitas kejadian stunting yang didapatkan dengan perhitungan berdasarkan teorema Bayes. Faktor determinan yang digunakan dalam pembuatan model prediksi meliputi berat badan lahir (kg), tinggi badan ibu (cm), tinggi ayah (cm), tingkat pendidikan ibu, status pekerjaan ibu, tempat tinggal, sanitasi, dan status ekonomi. Dari hasil tugas akhir ini ditemukan bahwa model prediksi Naive Bayes dengan performa terbaik didapatkan dengan proporsi pembagian data 90% - 10% dengan nilai SSE sebesar 73.9682 dan RMSE sebesar 0.446. Hasil dari ekstraksi informasi probabilitas menemukan bahwa semakin rendah tinggi ibu, tinggi ayah, tingkat pendidikan ibu, dan kuintil ekonomi keluarga maka semakin tinggi probabilitas kejadian anak stunting. Ibu yang tidak bekerja memiliki risiko probabilitas anak stunting lebih tinggi dibanding ibu yang bekerja. Keluarga yang bertempat tinggal di desa memiliki probabilitas anak stunting lebih tinggi dibanding keluarga yang bertempat tinggal di kota. Anak dengan berat lahir rendah memiliki probabilitas stunting lebih tinggi dibanding anak dengan berat lahir normal. Keluarga dengan sanitasi buruk memiliki risiko anak stunting lebih tinggi dibandingkan keluarga dengan sanitasi baik. =================================================================================================== Stunting is a condition where child under five years of age have a length or height that is less than their age. Stunting has a negative impact on the health of individuals who suffer from it in the short and long term. The national stunting prevalence of children under five years of age in Indonesia according to the 2018 Basic Health Research is 30.8%. This figure is higher than the world's stunting prevalence in 2019 according to the World Health Organization (WHO) of 21.3% and Southeast Asia of 24.7%. Therefore, efforts are needed to reduce the stunting rate in Indonesia. Research related to stunting in Indonesia is currently mostly in the form of analysis of the determinants that affect stunting and there are still few studies that discuss early prediction models of stunting risk and implement them in the form of applications that are easily accessible to the public. This final project aims to develop a web-based application to predict the risk probability of stunting in early childhood based on determinant factors. The probability prediction model is made using the Naive Bayes algorithm. This final project also aims to extract and disseminate information on the effect of each determinant on the probability of stunting obtained by calculations based on Bayes' theorem. The determinant factors used in making the prediction model include birth weight (kg), mother's height (cm), father's height (cm), mother's education level, mother's employment status, place of residence, sanitation, and economic status. From the results of this final project, it was found that the Naive Bayes prediction model with the best performance was obtained with a data partitioning proportion of 90% - 10% with an SSE value of 73.9682 and an RMSE of 0.446. The results of the extraction of probability information found that the lower the height of the mother, the height of the father, the education level of the mother, and the economic quintile of the family, the higher the probability of stunted children. Mothers who do not work have a higher probability of stunted child than mothers who work. Families who live in villages have a higher probability of stunted children than families who live in cities. Children with low birth weight have a higher probability of stunting than children with normal birth weight. Families with poor sanitation have a higher risk of stunted children than families with good sanitation.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Stunting, Prediksi Risiko, Teorema Bayes, Naive Bayes, Risk Prediction, Bayes Theorem
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > QA Mathematics > QA76.758 Software engineering
Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User: Abraham Christopher Pujiantoro
Date Deposited: 23 Aug 2021 03:36
Last Modified: 23 Aug 2021 03:36
URI: https://repository.its.ac.id/id/eprint/88745

Actions (login required)

View Item View Item