Nelvinebi/Water-Quality-Index-Prediction-for-Niger-Delta-Rivers-Using-ML-and-XGBoost

Domain:

environment and energy

Record type:

project

Creator:

Nel

Host:

This project leverages machine learning and XGBoost to predict the Water Quality Index of Niger Delta rivers using synthetic environmental data, supporting data-driven water quality assessment, pollution monitoring, and sustainable water resource management. ### 💧 Water Quality Index Prediction for Niger Delta Rivers Using ML & XGBoost > A machine learning pipeline that predicts **Water Quality Index (WQI)** for Niger Delta rivers using **XGBoost, Random Forest, and Linear Regression**, with **SHAP explainability** and **geospatial pollution mapping** to identify critical contamination zones across 4 states, 16 rivers, and 35 monitoring stations. --- ## 📌 Problem Water pollution in the **Niger Delta region** poses severe environmental and public health risks due to decades of oil exploration, industrial discharge, and inadequate wastewater management. Despite **16 major rivers** spanning **4 states**, comprehensive water quality monitoring remains fragmented, and accessible, analysis-ready datasets are scarce for researchers and policymakers. There is a critical need for **automated, reproducible ML pipelines** that can: - Calculate **Water Quality Index (WQI)** from multiple physicochemical parameters - Identify **pollution drivers** using explainable AI (SHAP) - Map **geospatial contamination patterns** across river zones - Provide **early warning systems** for hazardous water conditions --- ## 🎯 Objective - Engineer **WQI from 11 water quality parameters** following WHO standards - Train and compare **3 ML regression models** (XGBoost, Random Forest, Linear Regression) - Identify **most influential pollution drivers** using SHAP feature importance - Create **interactive geospatial pollution maps** (Folium) for stakeholder decision-making - Deploy a **Streamlit dashboard** for real-time WQI prediction and exploration - Provide a **reproducible framework** for water quality monitoring in developing regions --- ## 🗂️ Dataset The dataset contains **1,000 water samples** from **35 monitoring stations** across **16 rivers** in **4 Niger Delta states** (2019–2023). | Feature | Description | Unit | |---------|-------------|------| | `pH` | Acidity/alkalinity | — | | `Dissolved_Oxygen` | Oxygen a …

Visit

github.com

Nelvinebi/hybrid-ai-lstm-xgboost-interpretable-water-quality-prediction

A hybrid AI system combining LSTM, XGBoost, and K-Means to predict and interpret water quality using

Nelvinebi/Harmful-Algal-Bloom-Prediction-in-Niger-Delta-Coastal-Waters-Using-Remote-Sensing-and-ML

Machine learning–driven prediction of harmful algal blooms in Niger Delta coastal waters using synth

Nelvinebi/Sediment-Transport-and-Coastal-Morphodynamics-Modeling-in-the-Niger-Delta-Using-ML

Machine learning–based modeling of sediment transport and coastal morphodynamics in the Niger Delta

Nelvinebi/Shoreline-Erosion-and-Accretion-Analysis-in-the-Niger-Delta-Using-Satellite-Imagery-and-ML

This project applies machine learning to analyze shoreline erosion and accretion in the Niger Delta

Nelvinebi/Mangrove-Degradation-and-Health-Assessment-in-the-Niger-Delta-Using-NDVI-NDWI-and-ML

This project evaluates mangrove ecosystem health in the Niger Delta using NDVI and NDWI indices with

Nelvinebi/Climate-Change-Impact-on-Coastal-Ecosystems-in-the-Niger-Delta-Using-ML-Based-Models

Machine learning–driven assessment of climate change impacts on Niger Delta coastal ecosystems using