Scholarlink

<p>Potable water is essential for human and environmental health. Traditional water quality assessments are slow and uncertain. To improve water potability classification, we propose a hybrid deep learning model. The architecture combines Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) networks for enhanced temporal feature learning. In this paper, we utilized the Water Potability dataset from Kaggle and applied hot deck imputation to handle missing values. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied to the training data, enhancing the model’s ability to learn from underrepresented potable water samples. To improve model performance, the Relief Algorithm selected features and Bayesian Optimization (BO) to optimize the hyperparameters of GRU-LSTM. The dataset was 70% training, 15% validation, and 15% testing. The proposed GRU-LSTM model has been assessed with others like GRU, LSTM, ResNet-50, and VGG-19 deep learning models. The GRU-LSTM model exceeded all other models in distinguishing potable from non-potable water samples with an accuracy of 98.98%, F1-score of 98.98%, recall of 99.00%, precision of 99.00%, and AUC of 1.000. On the other hand, GRU had 95.12% accuracy, followed by LSTM with 94.30%, ResNet-50 with 91.46%, and VGG-19 with 84.96%, proving the hybrid model’s robustness. The results prove that the proposed methodology is used to classify water quality in real time with high reliability and efficiency. These findings demonstrate the potential of deep learning to enhance real-time environmental monitoring and inform water resource management.</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

Enhancing water potability classification using a hybrid GRU with LSTM deep learning models based on relief algorithm and bayesian optimization

  • 期刊论文

  • 发表日期：

    2025年12月22日

查看全文
  • Ahmed M. Elshewey,
  • Hazem M. El-Bakry,
  • Ahmed M. Osman

摘要

Potable water is essential for human and environmental health. Traditional water quality assessments are slow and uncertain. To improve water potability classification, we propose a hybrid deep learning model. The architecture combines Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) networks for enhanced temporal feature learning. In this paper, we utilized the Water Potability dataset from Kaggle and applied hot deck imputation to handle missing values. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied to the training data, enhancing the model’s ability to learn from underrepresented potable water samples. To improve model performance, the Relief Algorithm selected features and Bayesian Optimization (BO) to optimize the hyperparameters of GRU-LSTM. The dataset was 70% training, 15% validation, and 15% testing. The proposed GRU-LSTM model has been assessed with others like GRU, LSTM, ResNet-50, and VGG-19 deep learning models. The GRU-LSTM model exceeded all other models in distinguishing potable from non-potable water samples with an accuracy of 98.98%, F1-score of 98.98%, recall of 99.00%, precision of 99.00%, and AUC of 1.000. On the other hand, GRU had 95.12% accuracy, followed by LSTM with 94.30%, ResNet-50 with 91.46%, and VGG-19 with 84.96%, proving the hybrid model’s robustness. The results prove that the proposed methodology is used to classify water quality in real time with high reliability and efficiency. These findings demonstrate the potential of deep learning to enhance real-time environmental monitoring and inform water resource management.