Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Indonesian Journal of Artificial Intelligence and Data Mining
Susmanto, Susmanto
Unknown Affiliation
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Indonesian Journal of Artificial Intelligence and Data Mining

 1 
Exploring New Frontiers: XCEEMDAN, Bidirectional LSTM, Attention Mechanism, and Spline in Stock Price Forecasting Kelvin, Kelvin; Sinaga, Frans Mikael; Winardi, Sunaryo; Susmanto, Susmanto
Indonesian Journal of Artificial Intelligence and Data Mining Vol 7, No 2 (2024): September 2024
Publisher : Universitas Islam Negeri Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24014/ijaidm.v7i2.29649

Abstract

The Attention Mechanism is acknowledged as a machine learning method proficient in managing relationships within sequential data, surpassing traditional models in this regard. However, the unique characteristics of stock data, including substantial volatility, multidimensionality, and non-linear patterns, present challenges in attaining accurate forecasts of stock prices. This research aims to tackle these hurdles by enhancing a prior model through the incorporation of an Attention Mechanism, resulting in an enhanced model. The forecasted data are standardized and prepared for analysis before undergoing signal decomposition into high and low-frequency components. Subsequently, the Attention Mechanism processes the high-frequency signals. Evaluation entails comparing the performance of the proposed model with that of the previous model using identical parameters. The findings indicate that the proposed model achieves a reduced RMSE value of 0.5708777053 compared to the previous model's average RMSE value of 0.5823726212, indicating enhanced accuracy in stock price prediction. This approach is anticipated to make a substantial contribution to the advancement of more dependable and effective stock price prediction models, addressing the limitations of prior methodologies
Co-Authors Kelvin Kelvin Sinaga, Frans Mikael Winardi, Sunaryo