Random Forest and Logistic Regression algorithms for prediction of groundwater contamination using ammonia concentration

Abstract The present study aims to develop an efficient predictive model for groundwater contamination using Multivariate Logistic Regression (MLR) and Random Forest (RF) algorithms. Contamination by ammonia is recorded many authors at Sohag Governorate, Egypt attributed urban growth, agricultural, industrial activities. Thirty-two samples representing the Quaternary aquifer are collected analyzed major cations (Ca, Mg, Na), ammonia, nitrate, phosphate, heavy metals. Lead, magnesium, iron, zinc variables used test with which as index contamination. Spatial distribution maps statistical analyses show a strong correlation of lead magnesium whereas iron less correlation. For model, data divided into 70% training 30% testing subsets. performance evaluated classification reports, confusion matrix. Results (1) high RF accuracy 93% (2) MLR increased from 70 83% when “SOLVER” “C” parameters modified. helps identify contaminated zones area proved usefulness machine learning models prediction concentration.