Pengembangan Dual-Net iTransformer untuk Peramalan Deret Waktu

Ardi, Bagaskoro Kuncoro (2025) Pengembangan Dual-Net iTransformer untuk Peramalan Deret Waktu. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05111940000048-Undergraduate_Thesis.pdf Restricted to Repository staff only until 1 April 2027. Download (3MB) | Request a copy

Abstract

Peramalan deret waktu merupakan aspek penting dalam berbagai domain, seperti ekonomi, kesehatan, dan cuaca, di mana keputusan yang tepat sangat bergantung pada prediksi yang akurat. Dengan meningkatnya volume dan kompleksitas data, metode peramalan yang efisien menjadi semakin penting. Penelitian ini membahas tiga metode utama: Online Sequential Extreme Learning Machine (OS-ELM), Online Recurrent Extreme Learning Machine (OR-ELM), dan iTransformer (Inverted Transformer). OS-ELM menawarkan pembelajaran online yang cepat, sementara OR-ELM menangkap dinamika temporal dengan lebih baik. iTransformer, sebagai inovasi terbaru, meningkatkan interpretabilitas dan kemampuan pengolahan informasi multivariat. Penulis mengusulkan pengembangan Dual-Net iTransformer, yang menggabungkan kekuatan iTransformer dengan pendekatan dual network untuk meningkatkan akurasi dan efisiensi dalam peramalan deret waktu. Penelitian ini bertujuan untuk membandingkan performa ketiga metode tersebut dan mengeksplorasi kontribusi Dual-Net iTransformer dalam meningkatkan hasil peramalan.
================================================================================================================================
Time-series forecasting is a crucial aspect in various domains, such as economics, healthcare, and weather, where accurate predictions significantly impact decision-making. With the increasing volume and complexity of data, efficient forecasting methods have become increasingly important. This study discusses three primary methods: Online Sequential Extreme Learning Machine (OS-ELM), Online Recurrent Extreme Learning Machine (OR-ELM), and iTransformer (Inverted Transformer). OS-ELM offers rapid online learning capabilities, while OR-ELM effectively captures temporal dynamics. As a recent innovation, iTransformer enhances interpretability and the ability to process multivariate information. The author proposes the development of Dual-Net iTransformer, which combines the strengths of iTransformer with a dual network approach to improve accuracy and efficiency in time-series forecasting. This research aims to compare the performance of these three methods and explore the contribution of Dual-Net iTransformer in enhancing forecasting outcomes.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Peramalan deret waktu, OS-ELM, OR-ELM, iTransformer, Dual-Net iTransformer, pembelajaran online, Time-series forecasting, online learning
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Bagaskoro Kuncoro Ardi
Date Deposited:	05 Feb 2025 06:00
Last Modified:	05 Feb 2025 06:00
URI:	http://repository.its.ac.id/id/eprint/117802

Actions (login required)

View Item