Abadi, Rizki Trisna Rully (2021) Penerapan metode long short term memory dalam memprediksi jumlah kasus covid-19. Undergraduate thesis, Universitas Islam Negeri Maulana Malik Ibrahim.
|
Text (Fulltext)
17650082.pdf - Accepted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB) | Preview |
Abstract
INDONESIA:
Tingkat kasus positif harian COVID-19 di Indonesia mengalami kenaikan dan penurunan yang sangat bervariasi. Hal tersebut mempengaruhi penyedia layanan kesehatan dan pembuat regulasi dalam membuat kebijakan karena kondisi kasus harian yang bervariasi. Oleh karena itu pada penelitian kali ini akan melakukan prediksi terhadap jumlah kasus harian COVID-19 di Indonesia. Long Short Term Memory dipilih sebagai metode penelitian karena Long Short Term Memory dapat mempelajari data time series dan melakukan forecasting dengan cukup baik. Penelitian ini menggunakan data sekunder yang diambil dari kawalcovid19.id yang tersedia di portal http://sinta.ristekbrin.go.id/covid/. Data tersebut dibagi menjadi data latih dan data uji masing-masing sebesar 300 baris data dan 60 baris data dengan satu feature “kasus harian”. Hasil pengujian dari model yang telah dibangun pada proses training menunjukkan nilai MSE dan MAE masing-masing sebesar 0.03 dan 0.15 pada data normalisasi dan 2303320.93 dan 1268.98 pada data denormalisasi
ENGLISH:
The daily rate of positive cases of COVID-19 in Indonesia has varied widely. This affects health care providers and regulators in making policies because of the varying daily case conditions. Therefore, in this study, we will predict the number of daily cases of COVID-19 in Indonesia. Long Short Term Memory was chosen as the research method because Long Short Term Memory can study time series data and perform forecasting quite well. This study uses secondary data taken from kawalcovid19.id which is available on the http://sinta.ristekbrin.go.id/covid/ portal. The data is divided into training data and test data, each with 300 data lines and 60 data lines with one “daily case” feature. The test results of the model that has been built in the training process show the MSE and MAE values of 0.03 and 0.15 respectively on the normalized data and 2303320.93 and 1268.98 on the denormalized data.
Item Type: | Thesis (Undergraduate) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Supervisor: | Hanani, Ajib and Santoso, Irwan Budi | |||||||||
Contributors: |
|
|||||||||
Keywords: | LSTM; Prediksi; Kasus Harian COVID-19 | |||||||||
Subjects: | 08 INFORMATION AND COMPUTING SCIENCES > 0801 Artificial Intelligence and Image Processing > 080199 Artificial Intelligence and Image Processing not elsewhere classified | |||||||||
Departement: | Fakultas Sains dan Teknologi > Jurusan Teknik Informatika | |||||||||
Depositing User: | Rizki Trisna Rully Abadi | |||||||||
Date Deposited: | 28 Dec 2021 13:02 | |||||||||
Last Modified: | 28 Dec 2021 13:02 | |||||||||
URI: | http://etheses.uin-malang.ac.id/id/eprint/32662 |
Downloads
Downloads per month over past year
Actions (login required)
View Item |