A Scalable Machine Learning Framework for Hydrological Water Quality Monitoring Using Physicochemical and Microbial Parameters

Bhowmik P. N., Saini K., Sai Priya N. T., Anand P., Ateş B.

Water (Switzerland), cilt.17, sa.14, 2025 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 17 Sayı: 14
  • Basım Tarihi: 2025
  • Doi Numarası: 10.3390/w17142158
  • Dergi Adı: Water (Switzerland)
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Agricultural & Environmental Science Database, Aqualine, Aquatic Science & Fisheries Abstracts (ASFA), CAB Abstracts, Compendex, Environment Index, Food Science & Technology Abstracts, Geobase, INSPEC, Pollution Abstracts, Veterinary Science Database, Directory of Open Access Journals
  • Anahtar Kelimeler: classification and regression, environmental data analytics, feature importance analysis, machine learning (ML), water pollution monitoring, Water Quality Index (WQI)
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

Monitoring river water quality is essential for environmental sustainability and public health. This study proposes a machine learning (ML)-based framework to model, predict, and classify the Water Quality Index (WQI) using river water samples collected across India. The dataset includes eight physicochemical and microbial parameters: Temperature, pH, Dissolved Oxygen, Biological Oxygen Demand (BOD), Conductivity, Nitrate/Nitrite, Fecal Coliform, and Total Coliform. The WQI was calculated using weighted aggregation and categorized into Excellent, Good, Medium, and Poor classes. Regression and classification models—such as Linear Regression, Random Forest, Gradient Boosting, and Logistic Regression—were evaluated using MAE, RMSE, R2, Accuracy, Precision, Recall, and F1-score. Spatial mapping and exploratory data analysis were conducted to identify regional patterns. Feature importance (Gini and permutation-based) and error analysis enhanced interpretability. The framework achieved over 95% agreement with manual WQI classification, highlighting its effectiveness for real-time, scalable water quality monitoring and policy support.