Comparison of gradient boosted decision trees and random forest for groundwater potential mapping in Dholpur (Rajasthan), India

Abstract

In the drought prone district of Dholpur in Rajasthan, India, groundwater is a lifeline for its inhabitants. With population explosion and rapid urbanization, the groundwater is being critically over-exploited. Hence the current groundwater potential mapping study was undertaken to ascertain the areas that are more likely to yield a larger volume of groundwater against those areas that have poor groundwater potential and accordingly perpetuate the much needed damage control. Thematic layers for 14 groundwater influencing factors were considered for the study region, including elevation, slope, aspect, plan curvature, profile curvature, topographic wetness index (TWI), geology, soil, land use, normalized difference vegetation index (NDVI), surface temperature, precipitation, distance from roads, and distance from rivers. These were then subjected to an overlay operation, with the groundwater inventory which comprised of the locations of observational groundwater wells. The resulting geospatial database was then used to train two decision tree based ensemble models: gradient boosted decision trees (GBDT) and random forest (RF). The predictive performance of these models was then compared using various performance metrics such as area under curve (AUC) of receiver operating characteristics (ROC), sensitivity, accuracy, etc. It was found that GBDT (AUC: 0.79) outperformed RF (AUC: 0.71). The validated GBDT model was then used to construct the groundwater potential zonation map. The generated map showed that about 20.2% of the region has very high potential, while 22.6% has high potential to yield groundwater, and approximately 19.9-17.5% of the study region has very low to low groundwater potential.

Spatial Prediction of Groundwater Potentiality Mapping Using Machine Learning Algorithms

“…Here, the researchers were used the AUC-ROC (area under the curve) to estimate the prediction efficiency of the implemented models. This is a pectoral representation of model performance through a diagnostic test (Sachdeva and Kumar, 2020). The AUC indicates the model sensitivity by plotting the actual prediction of groundwater on the vertical axis (Y) and the false prediction of groundwater on the horizontal axis (X) (Gayen et al 2018;Gayen et al 2019).…”

“…A set of 15 × x15 toposheet maps were applied to calculate the stream junction density, pond density, and drainage density. Elevation is considered by many researchers (Golkarian et al, 2018;Sachdeva and Kumar, 2020) to be an important factor in GWPM. Elevation ranges from 13m to 120m in the Md.…”

“…Random forest is a well-known machine learning algorithm for both regression and classification tasks (Breiman, 2001;Naghibi et al 2017;Mandal and Pal, 2020). For regression and classification of the training datasets, researchers will attempt to develop a decision tree (DT) for class-wise output extraction and to acquire the dependent variable (Sachdeva and Kumar, 2020). This algorithm is structured by numerous DTs and composites to determine the relationship between the factors affecting groundwater and dug well occurrence (Kim et al 2018).…”

Determining residuary resistance per unit weight of displacement with Symbolic Regression and Gradient Boosted Tree algorithms

“…GBT is a tree-based AI ensemble method [39]. It is also based on trees such as SR, but instead of those trees describing equations, they describe decision paths [40].…”

Analysis of Data Mining Process for Improvement of Production Quality in Industrial Sector

“…The main notion is for computing the sequence of (very) simple trees, where each successive tree is established for the prediction residual of the preceding tree 18,19 . As in the General Classification and Regression Trees, this technique built on binary trees as an example, the partition of the data 20 into two samples at each split node. Now suppose that to limit the complexity of the trees to only three nodes: Two child nodes and a root node, i.e., a single split.…”