Scholarlink

<p>Climate change significantly impacts plant growth, food production, ecosystems, sustainable socio-economic development, and human health. With the growing availability of extensive historical climate data and the increasing demand for accurate production forecasting, there is a pressing need for reliable methods to determine the stochastic relationship between past and future values. This article introduces a novel deep learning model designed to overcome the limitations of traditional forecasting methods and achieve highly accurate predictions. The proposed approach is a deep long short-term memory (DLSTM) model optimized using genetic algorithms (GAs) and the mountain gazelle optimizer (MGO), which collectively fine-tune the architecture of the DLSTM model. The experiment utilized historical climate data from nine Egyptian cities: <b>Asswan</b>, <b>Bane-Suef</b>, <b>Behira</b>, <b>Dakhalia</b>, <b>Menoufia</b>, <b>Minia</b>, <b>Qalyubia</b>, <b>Sharkia</b>, and <b>Sohag</b> to evaluate the model’s performance. To ensure a fair and comprehensive evaluation, the effectiveness of the proposed MGO-GA-DLSTM model was compared with other established forecasting techniques. The evaluation metrics included <b>mean absolute error (MAE)</b>, <b>root mean square error (RMSE)</b>, <b>mean absolute percentage error (MAPE)</b>, and <b>R-squared </b>(<InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10462_2025_11247_Article_IEq1.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="21" /> </InlineMediaObject> <EquationSource Format="TEX">\(R^2\)</EquationSource> </InlineEquation>) to quantify prediction accuracy and model robustness. The findings demonstrate that the MGO-GA-DLSTM model outperforms existing methods in climate prediction, offering improved accuracy and reliability.</p>

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

Optimized deep learning architecture for predicting maximum temperatures in key Egyptian regions using hybrid genetic algorithm and mountain Gazelle optimizer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年6月4日

查看全文
  • Essam H. Houssein,
  • Mahmoud Dirar,
  • A. A. Khalil,
  • Abdelmaged A. Ali,
  • Waleed M. Mohamed

摘要

Climate change significantly impacts plant growth, food production, ecosystems, sustainable socio-economic development, and human health. With the growing availability of extensive historical climate data and the increasing demand for accurate production forecasting, there is a pressing need for reliable methods to determine the stochastic relationship between past and future values. This article introduces a novel deep learning model designed to overcome the limitations of traditional forecasting methods and achieve highly accurate predictions. The proposed approach is a deep long short-term memory (DLSTM) model optimized using genetic algorithms (GAs) and the mountain gazelle optimizer (MGO), which collectively fine-tune the architecture of the DLSTM model. The experiment utilized historical climate data from nine Egyptian cities: Asswan, Bane-Suef, Behira, Dakhalia, Menoufia, Minia, Qalyubia, Sharkia, and Sohag to evaluate the model’s performance. To ensure a fair and comprehensive evaluation, the effectiveness of the proposed MGO-GA-DLSTM model was compared with other established forecasting techniques. The evaluation metrics included mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE), and R-squared ( \(R^2\) ) to quantify prediction accuracy and model robustness. The findings demonstrate that the MGO-GA-DLSTM model outperforms existing methods in climate prediction, offering improved accuracy and reliability.