Saumen Maiti
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Saumen Maiti has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
Ram Krishna Tiwari
CSIR—National Geophysical Research Institute
4 shared publicationsGautam Gupta
Indian Institute of Geomagnetism (DST)
3 shared publicationsAmrita Singh
Indian Institute of Technology (Indian School of Mines)
2 shared publicationsAnasuya Das
Indian Institute of Technology (Indian School of Mines)
2 shared publicationsR. K. Tiwari
CSIR-National Geophysical Research Institute
2 shared publicationsBhabani Sankar Maity
Indian Institute of Technology (Indian School of Mines)
1 shared publication
Latest Publications
- 29 August 2017
- Published by Springer Nature in Modeling Earth Systems and Environment
- Vol. 3 (4) , 1371-1382
Application of automatic relevance determination model for groundwater quality index prediction by combining hydro-geochemical and geo-electrical data
- 29 August 2017
- Published by Springer Nature in Modeling Earth Systems and Environment
- Vol. 3 (4) , 1371-1382
Application of automatic relevance determination model for groundwater quality index prediction by combining hydro-geochemical and geo-electrical data
- 21 September 2016
- Published by Springer Nature in Stochastic Environmental Research and Risk Assessment
- Vol. 31 (7) , 1709-1726
Estimation of spatial variability of aquifer parameters from geophysical methods: a case study of Sindhudurg district, Maharashtra, India
- 21 September 2016
- Published by Springer Nature in Stochastic Environmental Research and Risk Assessment
- Vol. 31 (7) , 1709-1726
Estimation of spatial variability of aquifer parameters from geophysical methods: a case study of Sindhudurg district, Maharashtra, India
- 16 June 2016
- Published by Springer Nature in Journal of Earth System Science
- Vol. 125 (4) , 761-776
Modelling discontinuous well log signal to identify lithological boundaries via wavelet analysis: An example from KTB borehole data
- 16 June 2016
- Published by Springer Nature in Journal of Earth System Science
- Vol. 125 (4) , 761-776
Modelling discontinuous well log signal to identify lithological boundaries via wavelet analysis: An example from KTB borehole data
- 9 August 2013
- Published by Springer Nature in Environmental Earth Sciences
- Vol. 71 (7) , 3147-3160
A comparative study of artificial neural networks, Bayesian neural networks and adaptive neuro-fuzzy inference system in groundwater level prediction
- 9 August 2013
- Published by Springer Nature in Environmental Earth Sciences
- Vol. 71 (7) , 3147-3160
A comparative study of artificial neural networks, Bayesian neural networks and adaptive neuro-fuzzy inference system in groundwater level prediction
- 17 August 2012
- Published by Springer Nature in Environmental Monitoring and Assessment
- Vol. 185 (4) , 3445-3465
Assessment of groundwater quality: a fusion of geochemical and geophysical information via Bayesian neural networks
Deplorable quality of groundwater arising from saltwater intrusion, natural leaching and anthropogenic activities is one of the major concerns for the society. Assessment of groundwater quality is, therefore, a primary objective of scientific research. Here, we propose an artificial neural network-based method set in a Bayesian neural network (BNN) framework and employ it to assess groundwater quality. The approach is based on analyzing 36 water samples and inverting up to 85 Schlumberger vertical electrical sounding data. We constructed a priori model by suitably parameterizing geochemical and geophysical data collected from the western part of India. The posterior model (post-inversion) was estimated using the BNN learning procedure and global hybrid Monte Carlo/Markov Chain Monte Carlo optimization scheme. By suitable parameterization of geochemical and geophysical parameters, we simulated 1,500 training samples, out of which 50 % samples were used for training and remaining 50 % were used for validation and testing. We show that the trained model is able to classify validation and test samples with 85 % and 80 % accuracy respectively. Based on cross-correlation analysis and Gibb’s diagram of geochemical attributes, the groundwater qualities of the study area were classified into following three categories: “Very good”, “Good”, and “Unsuitable”. The BNN model-based results suggest that groundwater quality falls mostly in the range of “Good” to “Very good” except for some places near the Arabian Sea. The new modeling results powered by uncertainty and statistical analyses would provide useful constrain, which could be utilized in monitoring and assessment of the groundwater quality.
- 17 August 2012
- Published by Springer Nature in Environmental Monitoring and Assessment
- Vol. 185 (4) , 3445-3465
Assessment of groundwater quality: a fusion of geochemical and geophysical information via Bayesian neural networks
Deplorable quality of groundwater arising from saltwater intrusion, natural leaching and anthropogenic activities is one of the major concerns for the society. Assessment of groundwater quality is, therefore, a primary objective of scientific research. Here, we propose an artificial neural network-based method set in a Bayesian neural network (BNN) framework and employ it to assess groundwater quality. The approach is based on analyzing 36 water samples and inverting up to 85 Schlumberger vertical electrical sounding data. We constructed a priori model by suitably parameterizing geochemical and geophysical data collected from the western part of India. The posterior model (post-inversion) was estimated using the BNN learning procedure and global hybrid Monte Carlo/Markov Chain Monte Carlo optimization scheme. By suitable parameterization of geochemical and geophysical parameters, we simulated 1,500 training samples, out of which 50 % samples were used for training and remaining 50 % were used for validation and testing. We show that the trained model is able to classify validation and test samples with 85 % and 80 % accuracy respectively. Based on cross-correlation analysis and Gibb’s diagram of geochemical attributes, the groundwater qualities of the study area were classified into following three categories: “Very good”, “Good”, and “Unsuitable”. The BNN model-based results suggest that groundwater quality falls mostly in the range of “Good” to “Very good” except for some places near the Arabian Sea. The new modeling results powered by uncertainty and statistical analyses would provide useful constrain, which could be utilized in monitoring and assessment of the groundwater quality.
- 26 October 2010
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research
- Vol. 115 (B10)
Neural network modeling and an uncertainty analysis in Bayesian framework: A case study from the KTB borehole site
A new probabilistic approach based on the concept of Bayesian neural network (BNN) learning theory is proposed for decoding litho‐facies boundaries from well‐log data. We show that how a multi‐layer‐perceptron neural network model can be employed in Bayesian framework to classify changes in litho‐log successions. The method is then applied to the German Continental Deep Drilling Program (KTB) well‐log data for classification and uncertainty estimation in the litho‐facies boundaries. In this framework, a posteriori distribution of network parameter is estimated via the principle of Bayesian probabilistic theory, and an objective function is minimized following the scaled conjugate gradient optimization scheme. For the model development, we inflict a suitable criterion, which provides probabilistic information by emulating different combinations of synthetic data. Uncertainty in the relationship between the data and the model space is appropriately taken care by assuming a Gaussian a priori distribution of networks parameters (e.g., synaptic weights and biases). Prior to applying the new method to the real KTB data, we tested the proposed method on synthetic examples to examine the sensitivity of neural network hyperparameters in prediction. Within this framework, we examine stability and efficiency of this new probabilistic approach using different kinds of synthetic data assorted with different level of correlated noise. Our data analysis suggests that the designed network topology based on the Bayesian paradigm is steady up to nearly 40% correlated noise; however, adding more noise (∼50% or more) degrades the results. We perform uncertainty analyses on training, validation, and test data sets with and devoid of intrinsic noise by making the Gaussian approximation of the a posteriori distribution about the peak model. We present a standard deviation error‐map at the network output corresponding to the three types of the litho‐facies present over the entire litho‐section of the KTB. The comparisons of maximum a posteriori geological sections constructed here, based on the maximum a posteriori probability distribution, with the available geological information and the existing geophysical findings suggest that the BNN results reveal some additional finer details in the KTB borehole data at certain depths, which appears to be of some geological significance. We also demonstrate that the proposed BNN approach is superior to the conventional artificial neural network in terms of both avoiding “over‐fitting” and aiding uncertainty estimation, which are vital for meaningful interpretation of geophysical records. Our analyses demonstrate that the BNN‐based approach renders a robust means for the classification of complex changes in the litho‐facies successions and thus could provide a useful guide for understanding the crustal inhomogeneity and the structural discontinuity in many other tectonically complex regions.
- 26 October 2010
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research
- Vol. 115 (B10)
Neural network modeling and an uncertainty analysis in Bayesian framework: A case study from the KTB borehole site
A new probabilistic approach based on the concept of Bayesian neural network (BNN) learning theory is proposed for decoding litho‐facies boundaries from well‐log data. We show that how a multi‐layer‐perceptron neural network model can be employed in Bayesian framework to classify changes in litho‐log successions. The method is then applied to the German Continental Deep Drilling Program (KTB) well‐log data for classification and uncertainty estimation in the litho‐facies boundaries. In this framework, a posteriori distribution of network parameter is estimated via the principle of Bayesian probabilistic theory, and an objective function is minimized following the scaled conjugate gradient optimization scheme. For the model development, we inflict a suitable criterion, which provides probabilistic information by emulating different combinations of synthetic data. Uncertainty in the relationship between the data and the model space is appropriately taken care by assuming a Gaussian a priori distribution of networks parameters (e.g., synaptic weights and biases). Prior to applying the new method to the real KTB data, we tested the proposed method on synthetic examples to examine the sensitivity of neural network hyperparameters in prediction. Within this framework, we examine stability and efficiency of this new probabilistic approach using different kinds of synthetic data assorted with different level of correlated noise. Our data analysis suggests that the designed network topology based on the Bayesian paradigm is steady up to nearly 40% correlated noise; however, adding more noise (∼50% or more) degrades the results. We perform uncertainty analyses on training, validation, and test data sets with and devoid of intrinsic noise by making the Gaussian approximation of the a posteriori distribution about the peak model. We present a standard deviation error‐map at the network output corresponding to the three types of the litho‐facies present over the entire litho‐section of the KTB. The comparisons of maximum a posteriori geological sections constructed here, based on the maximum a posteriori probability distribution, with the available geological information and the existing geophysical findings suggest that the BNN results reveal some additional finer details in the KTB borehole data at certain depths, which appears to be of some geological significance. We also demonstrate that the proposed BNN approach is superior to the conventional artificial neural network in terms of both avoiding “over‐fitting” and aiding uncertainty estimation, which are vital for meaningful interpretation of geophysical records. Our analyses demonstrate that the BNN‐based approach renders a robust means for the classification of complex changes in the litho‐facies successions and thus could provide a useful guide for understanding the crustal inhomogeneity and the structural discontinuity in many other tectonically complex regions.
- 20 August 2009
- Published by Springer Nature in Pure and Applied Geophysics
- Vol. 166 (12) , 2059-2090
A Hybrid Monte Carlo Method Based Artificial Neural Networks Approach for Rock Boundaries Identification: A Case Study from the KTB Bore Hole
- 20 August 2009
- Published by Springer Nature in Pure and Applied Geophysics
- Vol. 166 (12) , 2059-2090
A Hybrid Monte Carlo Method Based Artificial Neural Networks Approach for Rock Boundaries Identification: A Case Study from the KTB Bore Hole
- 1 May 2007
- Published by Oxford University Press (OUP) in Geophysical Journal International
- Vol. 169 (2) , 733-746
Neural network modelling and classification of lithofacies using well log data: a case study from KTB borehole site
- 1 May 2007
- Published by Oxford University Press (OUP) in Geophysical Journal International
- Vol. 169 (2) , 733-746
Neural network modelling and classification of lithofacies using well log data: a case study from KTB borehole site