Analysis and Prediction of Groundwater Level ...

Analysis and Prediction of Groundwater Level using Machine Learning Models

Author Name : Narreddy Bharath Kumar Reddy, Manoj R, Pavan Kumar R, Niranjan N, Divya U H

ABSTRACT

Ground water is a critical resource in India and plays an essential role in meeting the water requirements of the country. However, with the increasing demand for water, depleting groundwater reserves have become a significant concern for the country. In our proposed system, we predict the annual ground water availability for future use by machine learning algorithms such as ridge regression, linear regression, random forest, decision tree, KNN, and SVM to predict the average ground water level for the years 2021 to 2025. The input data used for modelling is the groundwater level from 2010 to 2020. So accurate groundwater level prediction can help us better understand and manage our water resources for the benefit of all. At the beginning, we were expecting an accuracy of 90%, but after the completion of the project, it has been found that all the models are capable of making accurate predictions, with the linear regression algorithm performing the best with an accuracy of 100%. Groundwater level prediction can aid in the proper management of groundwater resources. Our study demonstrates the potential of machine learning models in predicting groundwater levels and highlights the importance of selecting appropriate input variables for model development.

Keywords—Groundwater level, Ridge regression, SVM, KNN, Logistic regression