India uses the most groundwater, with over 230 cubic kiloliters produced annually by its 27.5 million wells. Vertical artesian wells are dug in the surface’s subsurface aquifer to gather water for a number of uses. of objectives. Due to the rising need for artesian wells, more are being drilled each year. factors including decreased rainfall, water scarcity, and groundwater depletion etc. The motor is removed as the water dries, and the exterior is not adequately sealed or covered. The borewell’s diameter is big enough for a kid to fall into. into the interior of the at present disused or inoperative well. Apparently, one Sadly, more than 40 kids have fallen into wells since 2009, according to a report. portion of Rescue efforts have fallen short. The present study includes an analysis. of the Kaggle dataset derived from the drill holes. The purpose of this work is aims to create predictive models using basic machine learning techniques. ANN models, logistic regression, and bays to “forecasting existing equipment failure wells that will soon be reported and confiscated.” i.e., to identify non-operational bore wells that demand quick intervention. The effectiveness of the measures were used to assess the model. Good results from logistic regression performance in identifying deadly inactive wells.

Author (s) Details

K. V. Uma
Department of Information Technology, Thiagarajar College of Engineering, Madurai, India.

T. Janani Priya
Department of Information Technology, Thiagarajar College of Engineering, Madurai, India.

K. P. Sabari Priya
Department of Information Technology, Thiagarajar College of Engineering, Madurai, India.

Jennifer Priscilla Vincent
Department of Information Technology, Thiagarajar College of Engineering, Madurai, India.

View Book:  https://stm.bookpi.org/RDST-V9/article/view/7546

Keywords: Bore wells, machine learning algorithms, logistic regression, naïve bayes, K-NN, predictive model, performance metrics.