Search

The potential of advanced tree-based models and optimized deep learning algorithms to predict fluvial bedload transport was explored, identifying the most flexible and accurate algorithm, and the optimum set of readily available and reliable inputs. Using 926 datasets for 20 rivers, the performance of three groups of models was tested: (1) standalone tree-based models Alternating Model Tree (AMT) and Dual Perturb and Combine Tree (DPCT); (2) ensemble tree-based models Iterative Absolute Error Regression (IAER), ensembled with AMT and DPCT; and (3) optimized deep learning models Long Short-Term Memory (LSTM) and Recurrent Neural Network (RNN) ensembled with Grey Wolf Optimizer. Comparison of the predictive performance of the models with that of commonly used empirical equations and sensitivity analysis of the driving variables revealed that: (i) the coarse grain-size percentile D90 was the most effective variable in bedload transport prediction (where Dx is the xth percentile of the bed surface grain size distribution), followed by D84, D50, flow discharge, D16, and channel slope and width; (ii) all tree-based models and optimized deep learning algorithms displayed ‘very good’ or ‘good’ performance, outperforming empirical equations; and (iii) all algorithms performed best when all input parameters were used. Thus, a range of different input variable combinations must be considered in the optimization of these models. Overall, ensemble algorithms provided more accurate predictions of bedload transport than their standalone counterpart. In particular, the ensemble tree-based model IAER-AMT performed most strongly, displaying great potential to produce robust predictions of bedload transport in coarse-grained rivers based on a few readily available flow and channel variables.

The human brain is considered as a self-organizing system with self-similarities at various temporal and spatial scales called “fractals”. In this scale-free system, it is possible to decode the complexity of information processing using fractal behavior. For instance, the complexity of information processing in the brain can be evaluated by fractal dimensions (FDs). However, it is unclear how over-elaboration of information processing impacts the dimensionality of its fractal behavior. In this study, we hypothesized that FDs of electroencephalogram (EEG) in obsessive-compulsive disorder (OCD) should be higher than healthy controls (HCs) because of exaggeration of information processing mainly in the frontal regions. Therefore, a group of 39 OCDs (age: 34.76±8.22, 25 female, 3 left-handed) and 19 HCs (age: 31.94±8.22, 11 female, 1 left-handed) were recruited and their brain activities were recorded using a 19-channel EEG recorder in the eyes-open resting-state condition. Subsequently, fractal dimensions of the cleaned EEG data were calculated using Katz’s method in a frequency band-specific manner. After the test of normality, significant changes in the OCDs as compared to the HCs were calculated using a two-sample t-test. OCDs showed higher FDs in the frontal regions in all frequency bands as compared to HCs. Although, significant increases were only observed in the beta and lower gamma bands, mainly at the high beta. Interestingly, neurophysiological findings also show association with severity of obsessive behaviors. The results demonstrate that complexity of information processing in the brain follows an intimate nature of structural and functional impairments of the brain in OCD. [ABSTRACT FROM AUTHOR]