Peramalan Data Univariat Menggunakan Metode Long Short Term Memory
Abstract
Peramalan data univariat mengacu pada kegiatan meramalkan nilai pada data dengan satu variabel independen yang mungkin muncul di masa depan berdasarkan nilai-nilai yang ada di masa lalu. Penelitian ini bertujuan untuk memperoleh model yang dibangun menggunakan pendekatan deep learning jenis supervised learning yaitu metode Long Short Term Memory (LSTM) yang diterapkan pada data univariat. Metode LSTM merupakan pengembangan dari metode Recurrent Neural Network (RNN) dengan menambahkan 3 gate yang mampu memilih informasi yang dibutuhkan untuk pelatihan sel sehingga mampu mengurangi kemungkinan exploding gradients dan vanishing gradients. Model dibangun dengan input layer LSTM dengan unit sel dan output dense layer dengan tambahan hyperparameter tuning yang diset menggunakan optimizer, fungsi aktivasi  dan , dan nilai epoch. Performa model peramalan diuji menggunakan mean absolute percentage error (MAPE).
Copyright (c) 2023 Helma Syifa Izzadiana, Herlina Napitupulu, Firdaniza Firdaniza
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish articles in SisInfo : Jurnal Sistem Informasi dan Informatika agree to the following terms:
- Authors retain copyright of the article and grant the journal right of first publication with the work simultaneously licensed under a CC-BY-SA or The Creative Commons Attribution–ShareAlike License.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).