PERBANDINGAN ALGORITMA GRADIENT DESCENT DAN RESILIENT BACKPROPAGATION UNTUK MEMPREDIKSI DATA KUMULATIF KASUS COVID-19 DI INDONESIA

FAIZUL, AKROM (2023) PERBANDINGAN ALGORITMA GRADIENT DESCENT DAN RESILIENT BACKPROPAGATION UNTUK MEMPREDIKSI DATA KUMULATIF KASUS COVID-19 DI INDONESIA. Sarjana / Sarjana Terapan (S1/D4) thesis, Universitas Muhammadiyah Semarang.

Preview	Text 1. HALAMAN DEPAN.pdf Download (497kB) | Preview
Preview	Text 5. ABSTRAK.pdf Download (432kB) | Preview
Preview	Text 11. BAB 1.pdf Download (567kB) | Preview
Preview	Text 12. BAB 2.pdf Download (742kB) | Preview
	Text 13. BAB 3.pdf Restricted to Repository staff only Download (578kB) | Request a copy
	Text 14. BAB 4.pdf Restricted to Repository staff only Download (671kB) | Request a copy
Preview	Text 15. BAB 5.pdf Download (431kB) | Preview
Preview	Text 16. DAFTAR PUSTAKA.pdf Download (566kB) | Preview
	Text 18. HASIL OLAH DATA.pdf Restricted to Repository staff only Download (1MB) | Request a copy
	Text 19. FULL TEXT SKRIPSI.pdf Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

ABSTRAK Akrom, Faizul, 2021. Perbandingan Algoritma Pelatihan Gradient Descent dan Resilient Backpropagation untuk Memprediksi Data Kumulatif Kasus COVID-19 di Indonesia. Skripsi, Program Studi Statistika, Universitas Muhammadiyah Semarang, Pembimbing I: Indah Manfaati Nur, M.Si. II: Fathurokhman Fauzi, S.Si, M.Stat. Coronavirus Disease 2019 (COVID-19) merupakan penyakit menular yang disebabkan oleh Coronavirus jenis baru. Indonesia juga menjadi salah satu negara terdampak pandemi COVID-19. Sampai tanggal 6 Juni 2022 terjadi kenaikan jumlah kumulatif kasus COVID-19 setiap harinya. Meningkatnya jumlah korban virus ini ditambah dengan munculnya klaster penyebaran baru, menjadi pengingat untuk melakukan berbagai upaya untuk menekan laju penyebaran virus ini. Salah satu upaya yang dapat dilakukan adalah dengan melakukan pemodelan prediksi jumlah kumulatif kasus positif COVID-19. Pada penelitian ini akan dibandingkan algoritma Gradient Descent dengan algoritma resilient pada proses pembelajaran dalam kemampuan memprediksi jumlah kumulatif kasus COVID-19 di Indonesia. Hasil penelitian menunjukkan bahwa rata-rata jumlah total kasus COVID-19 yaitu sebesar 7.324 atau kurang lebih sebanyak 7.324 orang per harinya. Algoritma pelatihan jaringan saraf tiruan terbaik untuk memprediksi jumlah kumulatif kasus COVID-19 di Indonesia adalah Resilient Backpropagation (trainrp) karena mempunyai tingkat konvergensi yang cepat dan membutuhkan lebih sedikit iterasi (epoch). Jumlah kumulatif kasus COVID-19 di Indonesia masih mengalami peningkatan pada 7 hari ke depan. Kasus terendah berada pada taggal 7 Juni 2022 sebanyak 6.052.996 kasus dan kasus tertinggi pada 13 Juni 2022 sebanyak 6.054.828 kasus. Kata Kunci: COVID-19, Prediksi, Gradient Descent, Resilient Backpropagation   ABSTRACT Akrom, Faizul, 2021, Comparison of Gradient Descent and Resilient Backpropagation Algorithms to Predict Cumulative Data on COVID-19 Cases in Indonesia. Thesis, Statistics Study Program, Muhammadiyah University of Semarang, Supervisor I: Indah Manfaati Nur, M.Si. II: Fathurokhman Fauzi, S.Si, M.Stat. Coronavirus Disease2019 (COVID-19) is an infectious disease caused by a new type of Coronavirus. Indonesia is also one of the countries affected by the COVID-19 pandemic. Until Juni 6, 2022, there was an increase in the cumulative number of COVID-19 cases every day. The increasing number of victims of this virus coupled with the emergence of new clusters of spread, is a reminder to make various efforts to suppress the rate of spread of this virus. One of the efforts that can be done is by modeling the prediction of the cumulative number of positive COVID-19 cases. In this study, a Gradient Descent algorithm will be compared with a resilient algorithm in the learning process in the ability to predict the cumulative number of COVID-19 cases in Indonesia. The results showed that the average total number of COVID-19 cases was 7.324 or approximately 7.324 people per day. The best artificial neural network training algorithm to predict the cumulative number of COVID-19 cases in Indonesia is Resilient Backpropagation (trainrp) because it has a fast convergence rate and requires fewer iterations (epochs). The cumulative number of COVID-19 cases in Indonesia is still increasing in the next 14 days. The lowest cases were on Juni 6, 2022 as many as 6.052.996 the highest cases and cases on Juni 13,2022 were 6.054.828 case. Keywords: COVID-19, Prediction, Gradient Descent, Resilient Backpropagation

Actions (login required)

View Item