Scholarlink

This study explores the efficacy of Recurrent Neural Network (RNN) models—specifically Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional LSTM (Bi-LSTM+), and Bidirectional GRU (Bi-GRU+)— in predicting tide levels in the Davao Gulf. Using Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) as bench-marks, models were trained and tested with data sourced from the National Mapping and Resource Information Authority Hydrography Branch. Results highlight the GRU model’s superior performance, achieving an MSE of 70.46, MAE of 6.69, and RMSE of 8.39, surpassing other models. Despite additional feature incorporation, Bi-LSTM+ and Bi-GRU+ did not significantly enhance predictive capabilities. These findings underscore GRU’s accuracy and reliability in tide-level forecasting, suggesting future research focus on data quality enhancement and real-time monitoring system integration.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Davao Gulf Tide Level Forecasting with Recurrent Neural Network Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kent Cyril Bordios,
  • Francis Nathanael De Villena,
  • Kristian Moreno,
  • Ivy Kim Dit Machica

摘要

This study explores the efficacy of Recurrent Neural Network (RNN) models—specifically Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Bidirectional LSTM (Bi-LSTM+), and Bidirectional GRU (Bi-GRU+)— in predicting tide levels in the Davao Gulf. Using Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) as bench-marks, models were trained and tested with data sourced from the National Mapping and Resource Information Authority Hydrography Branch. Results highlight the GRU model’s superior performance, achieving an MSE of 70.46, MAE of 6.69, and RMSE of 8.39, surpassing other models. Despite additional feature incorporation, Bi-LSTM+ and Bi-GRU+ did not significantly enhance predictive capabilities. These findings underscore GRU’s accuracy and reliability in tide-level forecasting, suggesting future research focus on data quality enhancement and real-time monitoring system integration.