Pradipta, Bagas Hani (2021) Perancangan Sistem Prediktor Penambahan Kasus Harian Coronavirus Disease-19 (Covid-19) dengan Jaringan Syaraf Tiruan Recurrent Berdasarkan Variabel Cuaca: Studi Kasus Kota Surabaya. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Text
02311740000065-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (4MB) | Request a copy |
Abstract
Penyakit Coronavirus disease-19 (Covid-19) merupakan masalah serius yang dihadapi oleh negara – negara di dunia termasuk Indonesia saat ini. Penelitian mengenai Covid-19 menemukan dugaan bahwa Covid-19 memiliki hubungan dengan variabel cuaca seperti halnya penyakit SARS yang diakibatkan oleh satu strain virus yang sama. Maka dari itu, dilakukan penelitian perancangan sistem prediktor penambahan kasus harian Covid-19 dengan studi kasus pada kota Surabaya berdasarkan variabel eksternal yaitu data cuaca di kota Surabaya menggunakan algoritma jaringan syaraf tiruan (JST) recurrent mengacu pada hasil korelasi antara variabel cuaca dengan penambahan kasus Covid-19. Metode korelasi yang digunakan ialah korelasi spearman rank dan data yang digunakan ialah data cuaca dan kasus Covid-19 baik dalam bentuk delta kasus maupun kasus kumulatif dalam rentang waktu 1 tahun. Adapun variasi yang digunakan ialah kondisi data setelah pengolahan dan timestep jaringan syaraf tiruan recurrent. Hasil penelitian ini ialah ditemukan adanya korelasi data cuaca dengan penambahan kasus Covid-19 yang nantinya akan menjadi basis penentuan variabel input untuk masing-masing output. JST recurrent memiliki performa yang mampu untuk diaplikasikan dalam sistem predictor dengan nilai RMSE training yang bernilai kurang dari 1 pada semua variasi simulasi yang telah dilakukan, dan JST recurrent mampu untuk memprediksi Delta Konfirmasi dengan MAPE 16.2598%, Delta Meninggal dengan MAPE 32.6707% Konfirmasi Kumulatif dengan MAPE 0.2663%, Meninggal Kumulatif dengan MAPE 0.08322%, dan Aktif Kumulatif dengan MAPE 3.0884%. Namun tidak dapat memberikan prediksi yang baik pada Delta Aktif dengan MAPE yang diberikan ialah 115.1223%.
==================================================================================================
Coronavirus disease-19 (Covid-19) currently has been a serious global problem including Indonesia. Research about Covid-19 found out about the assumption that Covid-19 has tie with metereology variable like SARS that was caused by virus with same strain with Covid-19. Based from that, the research has been done to design predictor system for daily case growth of Covid-19 with case study in Surabaya using external variable which is metereology data in Surabaya using recurrent neural network based on the correlation between metereology variable and Covid-19 case. The correlation method used in this research is Spearman rank and the data used in this research is daily average of metereology data and daily case of Covid-19 in one-year range. The variation used in this research is data condition after being processed and the timestep in recurrent neural network. The result is there is a correlation between metereology data and Covid-19 case which would be basis for the input variable selection for each output. Recurrent neural network has good performace for the application of predictor system based on the training RMSE which has value less than 1 for every variation, and recurrent neural network is able to predict delta confirmation with MAPE 16.25975%, delta death with MAPE 32.67068%, cumulative confirmation with MAPE 0.266331%, cumulative death with MAPE 0.083218%, and cumulative active with MAPE 3.088425%. but the system can’t give good prediction to delta active with MAPE 115.1223%.
Item Type: | Thesis (Undergraduate) |
---|---|
Uncontrolled Keywords: | Sistem prediktor, Jaringan syaraf tiruan recurrent, Covid-19, Cuaca, Korelasi, Predictor system, recurrent neural network, Covid-19, Metereology, Correllation |
Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > T Technology (General) 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: | Bagas Hani Pradipta |
Date Deposited: | 16 Aug 2021 07:00 |
Last Modified: | 16 Aug 2021 07:00 |
URI: | http://repository.its.ac.id/id/eprint/87013 |
Actions (login required)
View Item |