Sürdürülebilir Kalkınma Amaçları
Yayınlar
- ArticleYalnız künye2025Irrigation water quality classification using boosting and CNN-BiGRU-Attention models: feature importance assessment using Hellinger distance
Zerouali, Bilel; Derdour, Abdessamed; Almalki, Abdulrahman Seraj; Katipoglu, Okan Mert; Arafat, Ahmed A.; Santos, Celso Augusto Guimarães; Bailek, Nadjem; Hussein, Enas E.
This study addresses a gap in irrigation water quality classification by comparing machine learning (ML) algorithms and deep learning (DL) models, with a focus on feature selection. Feature relevance was assessed using Hellinger distance, mutual information, and Chi-square analysis. Classifiers included boosting algorithms (AdaBoost, XGBoost, CatBoost, LightGBM, GradientBoosting, HistGradientBoosting), Extra Trees, and DL models (CNN-BiGRU and CNN-BiGRU-Attention). Performance was evaluated using standard metrics on 166 samples from Algeria’s Naama region. DL models achieved the highest accuracies (0.82–0.84) and AUC scores (0.98–0.981), followed by XGBoost and CatBoost. Feature analysis identified qiCl as most impactful (Hellinger distance = 0.248), and qiNa as most informative (mutual information= 1.0212). These findings underscore the strength of DL approaches and the value of integrated feature selection. The results offer practical insights for optimizing irrigation strategies, supporting risk assessment, and guiding sustainable water resource management.
RDL Kalıcı Kaydıİncele - ArticleYalnız künye2024Enhancing groundwater potential zone mapping with a hybrid analytical method: The case of semiarid basin
Zerouali, Bilel; Bailek, Nadjem; Islam, Abu Reza Md Towfiqul; Katipoğlu, Okan Mert; Ayek, Almustafa Abd Elkader; Santos, Celso Augusto Guimarães; Rajput, Jitendra; Wong, Yong Jie; Abda, Zaki; Chettih, Mohamed
Groundwater resource management is a critical component of sustainable water use, necessitating accurate and nuanced mapping of groundwater potential zones. This study analyzed the groundwater potential of Algeria's 43,750 km2 Chelif Basin (more than 17% of the area of Northern Algeria) using a combination of both subjective and objective mapping techniques. The adopted approaches encompass the benchmark analytical Fuzzy Analytic Hierarchy Process (Fuzzy-AHP) and the DEcision MAking Trial and Evaluation Laboratory (DEMATEL) methodologies to quantify interdependencies of criteria related to groundwater potential. The analysis focused on ten criteria related to groundwater potential, including core moisture availability and key hydrological factors like distance to river, topographic wetness index, and hydrological soils. Fuzzy-AHP achieved slightly higher groundwater prospecting accuracy (AUC = 0.730) than classic AHP (0.716), with benchmarking against 15 related studies indicating robust performance. Instead of the most commonly used criteria in groundwater literature such as lineament, stream order, recharge rate, and drainage density, this study employed alternative factors to challenge and validate the efficacy of the proposed methodology. The decision to omit certain criteria facilitated a more focused and manageable analysis, yet still delivered a robust evaluation of groundwater potential in the studied area. Moreover, this approach underscores the adaptability of the proposed methodology to accommodate varying sets of criteria, tailored according to the availability of data and specific research objectives. Additionally, the DEMATEL evaluation reveals new insights into subtle prioritization divergences, specifically between domain specialist opinions and analytical assessments of the criteria. The integration of fuzzy logic and causal relationship mapping through DEMATEL provides a comprehensive and robust foundation for groundwater potential modeling.
RDL Kalıcı Kaydıİncele - ArticleAçık Erişim2024Fine-tuning inflow prediction models: integrating optimization algorithms and TRMM data for enhanced accuracy
Ali, Enas; Zerouali, Bilel; Tariq, Aqil; Katipog, Okan Mert; Bailek, Nadjem; Santos, Celso Augusto Guimarães; Ghoneim, Sherif S.M.; Towfiqul Islam, Abu Reza Md.
This research explores machine learning algorithms for reservoir inflow prediction, including long short-term memory (LSTM), random forest (RF), and metaheuristic-optimized models. The impact of feature engineering techniques such as discrete wavelet transform (DWT) and XGBoost feature selection is investigated. LSTM shows promise, with LSTM-XGBoost exhibiting strong generalization from 179.81 m3/s RMSE (root mean square error) in training to 49.42 m3/s in testing. The RF-XGBoost and models incorporating DWT, like LSTM-DWT and RF-DWT, also perform well, underscoring the significance of feature engineering. Comparisons illustrate enhancements with DWT: LSTM and RF reduce training and testing RMSE substantially when using DWT. Metaheuristic models like MLP-ABC and LSSVR-PSO benefit from DWT as well, with the LSSVR-PSO-DWT model demonstrating excellent predictive accuracy, showing 133.97 m3/s RMSE in training and 47.08 m3/s RMSE in testing. This model synergistically combines LSSVR, PSO, and DWT, emerging as the top performers by effectively capturing intricate reservoir inflow patterns.
RDL Kalıcı Kaydıİncele