Prediction of Groundwater Level Using Machine Learning

Abstract

In this paper, Prediction of groundwater level is made from different states. In order to describe this information for the states and created seasonal models to depict groundwater activity, The developed models in this manner may be expanded for use by the individual remote sensing techniques created by the Division for Self-Organizing Cognitive Computing (CSOIS). When it came to minimising, Support vector regression consistently beat other techniques both root mean square error and mean absolute error. Additionally, Models that could be constructed using only local geographic information seemed to have more error than those that could be established with taking a global feature from a Gaussian Mixture Model.