Publication:
Modeling of standardized groundwater index of Bihar using machine learning techniques

dc.citedby10
dc.contributor.authorKumari S.en_US
dc.contributor.authorKumar D.en_US
dc.contributor.authorKumar M.en_US
dc.contributor.authorPande C.B.en_US
dc.contributor.authorid58164893400en_US
dc.contributor.authorid58248506100en_US
dc.contributor.authorid57713959100en_US
dc.contributor.authorid57193547008en_US
dc.date.accessioned2024-10-14T03:18:14Z
dc.date.available2024-10-14T03:18:14Z
dc.date.issued2023
dc.description.abstractGroundwater is the most preferred source of water resource for the human needs. Over-exploitation of the groundwater has been led to the tremendous effect on the groundwater drought. Assessment of groundwater drought is difficult to understand due to its complexity, non-linearity feature. In this study, groundwater drought indices of the state of Bihar, India, have been modeled using machine learning technique. The prediction of SGI was done by using Artificial Neural Network (ANN) and Random Forest (RF) machine learning models. The best input combinations of Gravity Recovery and Climate experiment (GRACE) satellite water equivalent data, rainfall data and groundwater level data was used to predicted the SGI. In this study, SGI of 38 districts of Bihar was calculated using groundwater data from 2002 to 2019. The accuracy and efficiency of the RF and ANN models were measured based on the mean square error (MSE) and correlation coefficient value (r). Compared to two models are shown the RF model is a performs superior as compare to ANN model, which model superior is decided based on the correlation coefficient value (r) as 0.95 and MSE value of 0.11. The results show both ML models such as ANN and RF is showed best results with the input combination of GRACE satellite water equivalent data, rainfall and groundwater data. � 2023 Elsevier Ltden_US
dc.description.natureFinalen_US
dc.identifier.ArtNo103395
dc.identifier.doi10.1016/j.pce.2023.103395
dc.identifier.scopus2-s2.0-85151276292
dc.identifier.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85151276292&doi=10.1016%2fj.pce.2023.103395&partnerID=40&md5=3b5fb50d66ed1793dba925536da9dfb2
dc.identifier.urihttps://irepository.uniten.edu.my/handle/123456789/34164
dc.identifier.volume130
dc.publisherElsevier Ltden_US
dc.sourceScopus
dc.sourcetitlePhysics and Chemistry of the Earth
dc.subjectArtificial neural network and random forest
dc.subjectGRACE satellite
dc.subjectGroundwater drought
dc.subjectStandardized groundwater index
dc.subjectBihar
dc.subjectIndia
dc.subjectDrought
dc.subjectForestry
dc.subjectGeodetic satellites
dc.subjectGroundwater resources
dc.subjectLearning algorithms
dc.subjectMachine learning
dc.subjectMean square error
dc.subjectNeural networks
dc.subjectRain
dc.subjectArtificial neural network and random forest
dc.subjectArtificial neural network modeling
dc.subjectCorrelation coefficient
dc.subjectGravity recovery and climate experiment satellites
dc.subjectGroundwater drought
dc.subjectMachine learning techniques
dc.subjectMeans square errors
dc.subjectRandom forests
dc.subjectStandardized groundwater index
dc.subjectWater equivalent
dc.subjectartificial neural network
dc.subjectGRACE
dc.subjectgroundwater
dc.subjectmodeling
dc.subjectrainfall
dc.subjectGroundwater
dc.titleModeling of standardized groundwater index of Bihar using machine learning techniquesen_US
dc.typeArticleen_US
dspace.entity.typePublication
Files
Collections