Please use this identifier to cite or link to this item:
http://repository.ipb.ac.id/handle/123456789/116042| Title: | Prediksi Harga Minyak Goreng Curah dan Kemasan Menggunakan Algoritme Long Short Term Memory(LSTM) |
| Authors: | Hasibuan, Lailan Sahrina Novialdi, Yanda |
| Issue Date: | 2023 |
| Publisher: | IPB University |
| Abstract: | Kenaikan harga bahan kebutuhan pokok yang signifikan akan menimbulkan
dampak negatif bagi perekonomian masyarakat Indonesia, seperti penurunan daya beli.
Berdasarkan monitoring yang dilakukan oleh Pusat Informasi Harga Pangan Strategis
sejak November 2021 hingga Agustus 2022 minyak goreng merupakan kebutuhan pokok
yang yang mengalami kenaikan harga secara signifikan dan terjadi merata di seluruh
indonesia, termasuk Jawa Barat. Dampak negatif dapat diatasi jika, fluktuasi harga dapat
diprediksi sebelumnya. Salah satu metode yang umum digunakan untuk prediksi fluktuasi
harga adalah LSTM. LSTM merupakan bagian dari deep Learning yang khusus
menangani data time series. Pada penelitian ini, penulis membangun model prediksi harga
minyak goreng curah dan kemasan pasa pasar tradisional di provinsi Jawa Barat
menggunakan metode Long Short-Term Memory (LSTM). Pemodelan prediksi harga
minyak goreng curah dan kemasan telah berhasil dibentuk dan memperoleh nilai yang
cukup baik untuk memprediksi harga di pasar tradisional. Berdasarkan beberapa
bercobaan yang dilakukan, hasil model prediksi terbaik untuk harga minyak goreng curah
dan kemasan diperoleh nilai NRMSE terkecil sebesar 0,01751 dan 0,03278. Berdasarkan
nilai NRMSE yang diperoleh, pembentukan model menggunakan LSTM menunjukan
hasil pada model mendekati variasi nilai aktualnya. A significant increase of price of basic necessities will have a negative impact on the economy of the Indonesian people, such as a decrease in purchasing power. Based on monitoring conducted by the Strategic Food Price Information Center from November 2021 to August 2022, cooking oil is a basic need that has experienced significant price increases and is occurring evenly throughout Indonesia, including West Java. Negative impacts can be overcome if price fluctuations can be predicted beforehand. One of the commonly used methods for predicting price fluctuations is LSTM. LSTM is part of deep learning that specifically handles time series data. In this study, the authors built a price prediction model for bulk cooking oil and packaging in traditional markets in West Java province using the Long Short-Term Memory (LSTM) method. Price prediction modeling for bulk and packaged cooking oil has been successfully established and obtained a good enough value to predict prices in traditional markets. Based on several experiments conducted, the best prediction model results for bulk and packaged cooking oil prices obtained the smallest NRMSE values of 0.01751 and 0.03278. Based on the NRMSE values obtained, the model formation using LSTM shows that the results on the model are close to the actual value variations. |
| URI: | http://repository.ipb.ac.id/handle/123456789/116042 |
| Appears in Collections: | UT - Computer Science |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Cover, Lembar Pengesahan, Prakata, Daftar Isi.pdf Restricted Access | Cover | 3.8 MB | Adobe PDF | View/Open |
| G64180037_Yanda Novialdi.pdf Restricted Access | Fullteks | 3.8 MB | Adobe PDF | View/Open |
| Lampiran.pdf Restricted Access | Lampiran | 3.8 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.