Intelligent Data Analytics Approaches for Predicting Dissolved Oxygen Concentration in River: Extremely Randomized Tree Versus Random Forest, MLPNN and MLR

Abstract

Control of water quality by monitoring water variables is still of major importance for the protection of human life (El Najjar et al. 2019). Rivers and streams are the major's component of the freshwater ecosystems and constitute the source of life for both humans and animals and they become the most "endangered ecosystems" in the world (Kumar and Jayakumar 2020; Kebede et al. 2020;Emenike et al. 2020). From year to year, a considerable work was carried out to maintain a good water quality status (Jerves-Cobo et al. 2020). Dissolved oxygen concentration (DO) in rivers, lakes and streams freshwater is a key variable in the assessment and control of the aquatic life and freshwater health (Banerjee et al. 2019). It also plays tremendous role in the control of water quality, and has many important effects on aquatic hydrochemistry and aquatic toxicology, and used as an "early warning" and classifying the eutrophication status of lakes (Hoang et al. 2019).Dissolved oxygen should be sufficiently and accurately estimated and monitored, and the variability of DO in freshwaters ecosystems can stem from various reasons and the amount of variability is particularly great when considering the interaction between several other water quality variables (Hutchins and Hitt 2019;Yaseen et al. 2018a). Over the years, it was demonstrated that there is substantial evidence that DO concentration is highly related to water quality variables (e.g. water temperature, water pH, specific conductance), which can be linked together. For example, it was demonstrated that decrease of DO concentration was mainly related to the increase S. Heddam (B) University 20 Août 1955 Skikda,

Predicting medical waste generation and associated factors using machine learning in the Kingdom of Bahrain

“…It is an algorithm that constructs an ensemble of un-pruned decision trees in a top-down fashion. The approach involves random cut-point selection for node splitting, rather than using bootstrap replicas, and utilizes the entire learning sample to grow the trees (Heddam, 2021). Moreover, this algorithm substantially reduces the prediction model's variance, which helps prevent overfitting, improves generalization performance, and slightly increases the model's bias, which may introduce a small trade-off between bias and variance.…”

Modeling Multistep Ahead Dissolved Oxygen Concentration Using Improved Support Vector Machines by a Hybrid Metaheuristic Algorithm

“…While recognizing that this modeling strategy has the advantage of having a simple and direct mathematical formulation, the poor generalization capabilities, in some cases, and the extensive research to determine the best model parameters using a hard trial-and-error process, alternative approaches that are based on model optimization using MOAs, can be seen as complementary alternative approaches that lead to the better selection of the model parameters. The literature review reveals that several single models for DO prediction are available in the literature, such as: the long short-term memory (LSTM) deep neural network model [14,15]; deterministic models (i.e., MINLAKE2018) [16]; the linear dynamic system and filtering model [17]; the gated recurrent unit (GRU) deep neural network model [18]; support vector regression (SVR) [19]; the stochastic vector auto regression (SVAR) model [20]; the radial basis function neural network (RBFNN) [21]; the multilayer perceptron neural network (MLPNN) [22]; polynomial chaos expansions (PCE) [23]; random forest regression (RFR) [24]; the extremely randomized tree [25]; and multivariate adaptive regression splines (MARS) [26].…”