Skripsi

PREDIKSI PENJUALAN MOBIL TOYOTA BERBASIS RECURRENT NEURAL NETWORK DENGAN ARSITEKTUR LONG SHORT-TERM MEMORY

---

Prediction is a process of estimating events that will occur in the future. In this research, software will be developed to predict Toyota car sales using the Long Short-Term Memory (LSTM) method, which is an improvement of the Recurrent Neural Network (RNN) method, to address the vanishing gradient problem when processing long-term sequential data. The data used in this study amounts to 149 data points, starting from January 2011 to May 2023. The model training in this research uses data split ratios of 90:10, 80:20, 70:30, and 60:40, each trained with parameters of 100, 200, and 300 epochs and a learning rate ranging from 10-1 to 10-4 to determine which configuration results in the lowest prediction error. The results of the study indicate that the model with a data split ratio of 90:10, 100 epochs, and a learning rate of 10-4 has the lowest prediction error among other model configurations, with an MSE value of 0.0152.

---

Availability

Detail Information

Series Title

-

Call Number

T1509742024

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2024

Collation

xvi, 153 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.320 7

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Long Short-Term Memory

Specific Detail Info

-

Statement of Responsibility

KA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment