Water Quality Prediction Model
Budget: ₹1,500 – ₹12,500 INR
This project aims to build a robust machine learning model to predict key water quality parameters across river locations. The training dataset includes geotagged measurements collected between 2011–2015 from ~200 sites.
The task is to enhance predictions by incorporating publicly available geospatial and environmental datasets such as Landsat Level‑2 satellite imagery, TerraClimate, or other open-source data sources. Beyond prediction accuracy, the project requires identifying key drivers of water quality variation through feature importance analysis.
Supporting material—including guidance documents, benchmark notebooks, and 2 datasets.
The task is to enhance predictions by incorporating publicly available geospatial and environmental datasets such as Landsat Level‑2 satellite imagery, TerraClimate, or other open-source data sources. Beyond prediction accuracy, the project requires identifying key drivers of water quality variation through feature importance analysis.
Supporting material—including guidance documents, benchmark notebooks, and 2 datasets.