Perancangan Sistem Prediktor dengan Jaringan Syaraf Tiruan (JST) Backpropagation untuk Menentukan Laju Pertambahan Kasus COVID-19 Berdasarkan Kadar Polutan PM2,5 : Studi Kasus Kota Surabaya

Asani, Putra Arisma Abdillah (2021) Perancangan Sistem Prediktor dengan Jaringan Syaraf Tiruan (JST) Backpropagation untuk Menentukan Laju Pertambahan Kasus COVID-19 Berdasarkan Kadar Polutan PM2,5 : Studi Kasus Kota Surabaya. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 02311740000002-Undergraduate_Thesis.pdf] Text
02311740000002-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (6MB) | Request a copy

Abstract

Kasus Covid-19 adalah salah satu penyakit yang saat ini telah menewaskan ratusan juta manusia di seluruh negara didunia. Peningkatan kasus Covid-19 ini masih terus meningkat khususnya di Indonesia. Pada penelitian kali ini bertujuan untuk membuat suatu sistem prediktor yang dapat memprediksi laju pertambahan Covid-19 berdasarkan variabel kadar polutan PM2,5 di kota Surabaya. Metode yang digunakan dalam sistem prediktor kali ini yaitu Jaringan Syaraf Tiruan (JST) backpropagation. Sistem prediktor ini sebelum masuk pada perancangan, pada penelitian ini dilakukan analisis korelasi terlebih dahulu antara kasus Covid-19 dengan kadar polutan PM2,5, temperatur dan kelembaban, metode yang digunakan yaitu analisis spearman rank. Pengujian dilakukan dibantu dengan software matlab untuk sistem prediktor dan SPSS untuk uji korelasi. Pada uji korelasi antara kasus pasien terkonfirmasi dan aktif terhadap kadar polutan didapatkan hasil yaitu -0,561 dan 0,496, sedangkan untuk kadar polutan terhadap temperatur dan kelembaban yaitu -0,242 dan -0,326. Sistem prediktor ini pada pengolahan data dibagi menjadi dua yaitu data latuh dan data uji, kemudian arsitektur dirancang untuk banyak percobaan dengan menetapkan jumlah hidden layer, epoch, maximum error (0 dan 0,001), dan learning rate. Model arsitektur terbaik yang didapatkan ada 3 yaitu dengan MSE 0,00099999, 0,001 dan 0,001 yang diperoleh dari rancangan arsitektur dengan 5 input dan 1 output untuk prediksi t+1 kasus terkonfirmasi.
====================================================================================================
Covid-19 is one of the diseases that currently has killed hundreds of millions people in all countries in the world. The Case of Covid-19 is still increasing, especially in Indonesia. The aim of this study is to create a predictor system that can predict the rate of increase in Covid-19 based on the variable PM2.5 pollutant levels in the city of Surabaya. The method that used in this predictor system was backpropagation neural network. Before making a predictor system, in this study, a correlation analysis was carried out between Covid-19 cases and PM2.5 pollutant levels, temperature and humidity, using Spearman rank analysis. The test was assisted by Matlab software for predictor systems and SPSS to make correlation tests. The result of correlation test between confirmed and active patient cases on pollutant levels were -0.561 and 0.496, while for pollutant levels to temperature and humidity were -0.242 and -0.326. In the predictor system, data processing was divided into two data. They are sinking data and test data, then the architecture was designed for many experiments by setting number of hidden layers, epochs, maximum error (0 and 0.001), and learning rate. There are three best architectural models obtained. They are MSE 0.00099999, 0.001 and 0.001 that obtained from architectural designs with 5 inputs and 1 output to predict t + 1 confirmed cases.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Covid-19, PM2,5, Backpropagation, Korelasi, MSE, Covid-19, Correlation, PM2.5, Backpropagation, MSE
Subjects: Q Science > Q Science (General) > Q325.78 Back propagation
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
T Technology > T Technology (General) > T174 Technological forecasting
Divisions: Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User: Putra Arisma Abdillah Asani
Date Deposited: 18 Aug 2021 14:53
Last Modified: 18 Aug 2021 14:53
URI: http://repository.its.ac.id/id/eprint/87415

Actions (login required)

View Item View Item