Search

The widespread deployment of deep nets in practical applications has lead to a growing desire to understand how and why such black-box methods perform prediction. Much work has focused on understanding what part of the input pattern (an image, say) is responsible for a particular class being predicted, and how the input may be manipulated to predict a different class. We focus instead on understanding which of the internal features computed by the neural net are responsible for a particular class. We achieve this by mimicking part of the neural net with an oblique decision tree having sparse weight vectors at the decision nodes. Using the recently proposed Tree Alternating Optimization (TAO) algorithm, we are able to learn trees that are both highly accurate and interpretable. Such trees can faithfully mimic the part of the neural net they replaced, and hence they can provide insights into the deep net black box. Further, we show we can easily manipulate the neural net features in order to make the net predict, or not predict, a given class, thus showing that it is possible to carry out adversarial attacks at the level of the features. These insights and manipulations apply globally to the entire training and test set, not just at a local (single-instance) level. We demonstrate this robustly in the MNIST and ImageNet datasets with LeNet5 and VGG networks. [ABSTRACT FROM AUTHOR]

The deepest all-sky survey available in the γ-ray band – the last release of the Fermi-LAT catalogue (4FGL-DR3) based on the data accumulated in 12 years contains more than 6600 sources. The largest population among the sources is blazar subclass – 3743, 60.1 per cent of which are classified as BL Lacertae objects (BL Lacs) or Flat Spectrum Radio Quasars (FSRQs), while the rest are listed as blazar candidates of uncertain type (BCU) as their firm optical classification is lacking. The goal of this study is to classify BCUs using different machine learning algorithms, which are trained on the spectral and temporal properties of already classified BL Lacs and FSRQs. Artificial Neural Networks, XGBoost , and LightGBM algorithms are employed to construct predictive models for BCU classification. Using 18 input parameters of 2219 BL Lacs and FSRQs, we train (80 per cent of the sample) and test (20 per cent) these algorithms and find that LightGBM model, state-of-the-art classification algorithm based on gradient boosting decision trees, provides the highest performance. Based on our best model, we classify 825 BCUs as BL Lac candidates and 405 as FSRQ candidates, however, 190 remain without a clear prediction, but the percentage of BCUs in 4FGL is reduced to 5.1 per cent. The γ-ray photon index, synchrotron peak frequency, and high-energy peak frequency of a large sample are used to investigate the relationship between FSRQs and BL Lacs (LBLs, IBLs, and HBLs). [ABSTRACT FROM AUTHOR]

This study addresses the limitations of current parametric equations and artificial neural networks (ANNs) in accurately predicting the stress concentration factor (SCF) of fillet welded joints stemming from the simplification of their real weld profiles. To improve the accuracy, this study introduces bagged trees for estimating local stresses. The dataset used as the foundation for training the bagged trees is extracted from the actual weld geometry of T-shaped joints. It is created via a digitalization process involving the extraction of actual geometric parameters from the joints, which are transformed into finite element models (FEMs). These models are then employed to determine the ratio between the simulated sectional stress and the nominal stress (σ sec / ∆ σ nom) under an axial loading condition. A comprehensive comparison is carried out among existing parametric equations, ANNs, and the proposed bagged trees. The results emphasize the inadequacy of idealized geometry models in accurately determining local stresses for real weld profiles. In contrast, bagged trees are a promising method for accurately computing sectional weld stresses (σ sec) within real weld geometry. • Considering the actual weld geometry to assess local stresses. • Using artificial intelligence to predict local stresses in welded joints. • Examining the applicability of developed methods for predicting the stress concentration factor considering the actual weld geometry. [ABSTRACT FROM AUTHOR]

A production plan gives the quantity of products to release on the shop floor in each period, where a period may represent a week or a month. The plan is the basis for negotiating order acceptance and delivery dates with customers and suppliers. The production plan must respect the available capacity on the shop floor, as underloading the shop floor leads to a loss of opportunity, and thus, a loss of competitiveness for the company. To properly manage the production capacity while negotiating with suppliers and customers, the production planners need a tool to accurately estimate the capacity consumption in each period. The computation of capacity consumption requires creating a detailed production schedule which is a complex task. Algorithms that find close to optimal schedules in a complex manufacturing environment are often time-consuming, which is impractical in a negotiation context. We investigate machine learning models to predict capacity consumption. We consider a flexible job-shop as commonly encountered in practice, and proposed several machine learning models. Namely, several variants of linear regression, decision trees, and artificial neural networks. Numerical experiments showed that our models outperform those found by both an exact approach and dispatching rules when computation time is short. [ABSTRACT FROM AUTHOR]

Simple Summary: In the management of beef cattle, it can be useful to divide individuals based on a specific trait value (below and above average). This in turn allows for focusing on a larger group of animals with the aim of improving, e.g., their growth rate or obtaining a more uniform group in terms of a given trait. Classifying calves into less (below average) and more (above average) efficient growth creates an opportunity for producers to direct their efforts towards the "worse" animals and improve their performance through adjustments in nutrition, animal grouping, or reorganization of work. In this study, models were developed based on data from a beef farm. They were used to classify beef calves into poorer and better growth groups. In order to obtain more input data, predictions were made for the third calf. Among the analyzed models, random forest was the most effective. The most significant factors influencing daily body weight gains were also identified and discussed in the present study. The results demonstrate that machine learning models can be useful for classifying calves based on their growth rates. However, it is necessary to maintain proper breeding documentation from which the predictors can be obtained. The aim of the present study was to compare the predictive performance of decision trees, artificial neural networks, and logistic regression used for the classification of daily body weight gains in beef calves. A total of 680 pure-breed Simmental and 373 Limousin cows from the largest farm in the West Pomeranian Province, whose calves were fattened between 2014 and 2016, were included in the study. Pre-weaning daily body weight gains were divided into two categories: A—equal to or lower than the weighted mean for each breed and sex and B—higher than the mean. Models were developed separately for each breed. Sensitivity, specificity, accuracy, and area under the curve on a test set for the best model (random forest) were 0.83, 0.67, 0.76, and 0.82 and 0.68, 0.86, 0.78, and 0.81 for the Limousin and Simmental breeds, respectively. The most important predictors were daily weight gains of the dam when she was a calf, daily weight gains of the first calf, sex of the third calf, milk yield at first lactation, birth weight of the third calf, dam birth weight, dam hip height, and second calving season. The selected machine learning models can be used quite effectively for the classification of calves based on their daily weight gains. [ABSTRACT FROM AUTHOR]

In recent years, reinforcement learning (RL) techniques have achieved great success in many different applications. However, their heavy reliance on complex deep neural networks makes most RL models uninterpretable, limiting their application in domains where trust and security are important. To address this challenge, we propose MENS-DT-RL, an algorithm capable of constructing interpretable models for RL via the evolution of decision tree (DT) models. MENS-DT-RL uses a multi-method ensemble algorithm to evolve univariate DTs, guiding the process with a fitness metric that prioritizes interpretability and consistent high performance. Three different initializations for the MENS-DT-RL are proposed, including the use of Imitation Learning (IL) techniques, and a novel pruning approach that reduces solution size without compromising performance. To evaluate the proposed approach, we compare it with other models from the literature on three benchmark tasks from the OpenAI Gym library, as well as on a fertilization problem inspired by real-world crop management. To the best of our knowledge, the proposed scheme is the first to solve the Lunar Lander benchmark with both interpretability and a high confidence rate (90% of episodes are successful), as well as the first to solve the Mountain Car environment with a tree of only 7 nodes. On the real-world task, the proposed MENS-DT-RL is able to produce solutions with the same quality as deep RL policies, with the added bonus of interpretability. We also analyze the best solutions found by the algorithm and show that they are not only interpretable but also diverse in their behavior, empowering the end user with the choice of which model to apply. Overall, the findings show that the proposed approach is capable of producing high-quality transparent models for RL, achieving interpretability without losing performance. • A multi-method ensemble evolutionary algorithm is proposed for Interpretable RL. • Interpretability is achieved by using Decision Trees as RL agents. • Imitation Learning is leveraged to jumpstart the evolutionary algorithm. • A novel method is proposed to prune DTs in RL environments. • State-of-the-art results are obtained for several RL benchmarks. [ABSTRACT FROM AUTHOR]

The majority of people are using connecting media in recent times, like Facebook, Twitter, LinkedIn, etc. People like to communicate on social websites as compared to direct conversation due to the lack of time, fast communication, minimizing long distance, etc. but more fraud such as fake accounts, cloned attacks, etc. are generated due to growing data on social connecting media. In our proposed approach, we are using the Bernoulli RBM_RF Classifier (BRRC) to detect cloned attacks in connecting media. This model is based on deep learning algorithms and more research is under process but still, there is no such kind of model to provide high accuracy. In this model, we have used multiple algorithms Artificial Neural Networks (ANN), Logistic Regression, Decision Trees, K-Nearest Neighbour Classification, Random Forest, and Bernoulli RBM with extra add-on features of the random forest we have created a Bernoulli pipeline also known as BRRC. Our goal is to establish a clear neural network connection between our target attributes. Finally, we created a restricted Boltzmann machine using a RBM Bernoulli pipeline with additional add-on random forest features after using random forest, where we employed n number of decision trees for classification and provided the same in depth. Our model can extract higher-level features from the given data and helps in the dimensionality Reduction of the input data by learning a compressed representation. Accuracy can be improved by using a big dataset with particular connecting media attributes that are either not currently available or are only sparingly present. We can conclude that Bernoulli Restricted Boltzmann Machine is the best fit and suitable for our model with 93% accuracy. [ABSTRACT FROM AUTHOR]

This paper proposes a novel ensemble approach, integrating Artificial Neural Networks (ANNs), Symbolic Regression (SR), and Decision Trees (DTs), to predict fish mortality caused by infectious diseases. The intensifying global burden of fish diseases threatens the sustainability of aquatic ecosystems and the aquaculture industry, necessitating sophisticated modelling strategies for effective disease management and control. The proposed approach capitalizes on the non-linear data modelling strength of ANNs, the explanatory power of SR, and the decision-making efficiency of DTs, offering both predictive accuracy and interpretable insights. The architecture of the proposed ensemble method is developed in two stages. In the intermediate stage, an ANN is employed to learn the complex, non-linear interactions between various biological and environmental factors impacting fish health. Additionally, SR is applied to produce a symbolic equation that effectively maps the input variables to fish mortality rates. In the final stage, a DT model is included to enhance prediction performance by capturing decision rules from the data. This hybrid approach offers superior prediction performance while also revealing meaningful biological/environmental relationships that can guide preventive and reactive interventions in the management of fish health. We evaluate the developed models using extensive real-world datasets acquired from two large Greek fish-farming units, which encompass representative disease types. The results demonstrate that our ensemble approach significantly outperforms traditional standalone models developed in our recent previous work, achieving enhanced predictive accuracy, robustness, and interpretability. Overall, this research has far-reaching implications for improving disease predictions, facilitating optimal decision-making in aquaculture management, and contributing to the sustainability of global fish stocks. [ABSTRACT FROM AUTHOR]

With the increased use of the Internet, unauthorized access has increased, allowing malicious users to hack networks and carry out malicious activities. One of the essential modern approaches in today's cybersecurity efforts is the limitation of access by suspect users. In this study, the approach toward real-time intrusion detection was to consider behavioral patterns of past users on the network. We classified the users as two categories: intervention and non-intervention, and employed the machine learning techniques Artificial Neural Networks [ANN], Support Vector Machines [SVM], and Decision Trees [DT]. The Decision Trees model was chosen as it had a mature capability concerning complex pattern recognition and an enhancement capability of the intrusion detection systems. The efficiency of these algorithms is examined via the key performance metrics: confusion matrix, F1-score, and Area under the Curve [AUC]. Decision Tree, which came up as the best model for both the training and testing phases, produced an outstanding F1-score of 99.96% and AUC score of 99.93% in the testing phase. SVM and ANN gave good results; the F1 scores of SVM and ANN in the testing phase were 92.76% and 93.33%, while the AUC was 90.57% and 94.78%, respectively. This research will enlighten us on the influence of machine learning on the scope of intrusion detection, fostering more development efforts toward more responsive and dynamic intrusion detection systems. The comparative evaluation of these models will help in providing vital information for the further enhancement of cybersecurity strategies, ensuring better defenses against these ever-evolving cyber threats. [ABSTRACT FROM AUTHOR]

Objective: This study aimed to evaluate the usability of morphometric features obtained from mandibular panoramic radiographs in gender determination using machine learning algorithms. Materials and methods: High-resolution radiographs of 200 patients aged 20–77 (41.0 ± 12.7) were included in the study. Twelve different morphometric measurements were extracted from each digital panoramic radiography included in the study. These measurements were used as features in the machine learning phase in which six different machine learning algorithms were used (k-nearest neighbor, decision trees, support vector machines, naive Bayes, linear discrimination analysis, and neural networks). To evaluate the reliability, we have performed tenfold cross-validation and we repeated this 10 times for every classification process. This process enhances the reliability of the results for other datasets. Results: When all 12 features are used together, the accuracy rate is found to be 82.6 ± 0.5%. The classification accuracies are also compared using each feature alone. Three features that give the highest accuracy are coronoid height (80.9 ± 0.9%), condyle height (78.2 ± 0.5%), and ramus height (77.2 ± 0.4%), respectively. When compared to the classification algorithms, the highest accuracy was obtained with the naive Bayes algorithm with a rate of 84.0 ± 0.4%. Conclusion: Machine learning techniques can accurately determine gender by analyzing mandibular morphometric structures from digital panoramic radiographs. The most precise results are achieved by evaluating the structures in combination, using attributes obtained from applying the MRMR algorithm to all features. [ABSTRACT FROM AUTHOR]

Machine learning is frequently used in various geotechnical applications nowadays. This study presents a statistics and machine learning model for settlement prediction of helical piles that relates compressive service load and soil parameters as a group with the pile parameters. Machine learning algorithms such as Decision Trees, Random Forests, AdaBoost, and Artificial Neural Networks (ANN) were used to develop the predictive models. The models were validated using cross-validation techniques and tested on an independent dataset to assess their accuracy and generalizability. Numerical investigation is used here to supplement the field data by simulating various soil conditions and pile geometries that have not been tested in the field. This study compiled numerical results of 3600 models. As the models are well-calibrated and validated, the data from these models can be reasonably assumed to simulate the ground situation. At the end of this study, a comparative analysis of statistic learning and machine learning (ML) was done using the field axial load tests database and numerical investigation on helical piles. It is observed that ML models like Decision Trees and Random Forests provided the better model with R-squared values of 0.92 and 0.96, respectively, for large diameters. The authors believe this study will permit engineers and state agencies to understand this prediction model's efficacy better, resulting in a more resilient approach to designing large-diameter helical piles for the compressive load. [ABSTRACT FROM AUTHOR]

• Sugar beet nitrogen fertilization was optimized using leaf nutrient status samples. • Root yield and yield quality parameters were used as training and test data. • Ensemble decision trees and artificial neural network regression were used. • Na had high variable importance for root yield and sucrose content. • The top accuracy had a median R2 of 0.927 and ranged from 0.842 to 0.998. Nitrogen fertilization has a crucial role in sugar beet production, especially concerning root yield and quality. This study employed a machine learning approach to predict root yield and quality parameters based on the nutrient status of sugar beet leaves in relation to nitrogen fertilization. The field experiment included the following N fertilization treatments for sugar beet production: control (N0), presowing (N1 = 45 kg N ha−1) and presowing with top-dressing (N2 = 99 and 154.5 kg ha−1). Leaf samples were collected during the vegetation period in six intervals (May-Sept) to determine the levels of N, K and Na in the leaf dry matter. The machine learning regression based on ensemble decision trees and artificial neural network was used to determine the relationship of leaf samples based on varying N fertilization with yield parameters. Among the leaf elements analyzed, Na exhibited the highest average relative variable importance for root yield, sucrose content, and other quality parameters during the season with greater precipitation. In the season with less precipitation, N content at the beginning of July showed higher importance on root yield (74.6). The evaluated machine learning methods consistently achieved high accuracy across various combinations of input data and yield parameters, with a median R2 of 0.927 and a range from 0.842 to 0.998. [ABSTRACT FROM AUTHOR]

Open channel bends have fascinated engineers and scientists for decades while providing water for domestic, irrigation and industrial consumption. The presence of curvature in a channel impacts the flow pattern, velocity and water surface profile. Simulating flow variables such as velocity and water surface depth is one of the most important matters in the design and application of open channel bends. This study investigates a new neural network method using the radial basis function (RBF) based on decision trees (DT-RBF) to predict velocity and free-surface water profiles in a 90° open channel bend. In this study, 506 flow depth and 520 depth-averaged velocity field data obtained at 5 different discharges (5, 7.8, 13.6, 19.1 and 25.3 l/s) in a 90° sharp bend were used for training and testing purposes. The obtained results showed that the proposed DT-RBF models were more accurate than RBF models in estimating flow depth and depth-averaged velocity in the bend. The RBF root-mean-square error (RMSE), mean absolute error (MAE) and relative error (δ) were reduced by 20, 24 and 23.5%, respectively, when using the hybrid DT-RBF model to estimate the depth-averaged velocity. For water surface prediction, the RMSE, MAE and δ decreased by 33, 27.5 and 37%, respectively, when using the proposed DT-RBF hybrid model. For the longitudinal profiles of water surface profile prediction at the outer edge, MAE (0.018) improved to MAE (0.0084) with DT-RBF. It was found that the hybrid decision tree-based method significantly improved RBF neural network performance in forecasting the velocity and free-surface water profiles in a 90° open channel sharp bend. [ABSTRACT FROM AUTHOR]

Using the complementary strengths of Bayesian networks, decision trees, artificial neural networks (ANNs), and Markov models, this endeavor intends to completely revamp clinical decision-making. In order to provide instantaneous access to image-driven insights and clinical decision support systems (CDSS), want to create a revolutionary framework that merges these cutting-edge methods with cloud-enabled technologies. The proposed framework gives a comprehensive perspective of patient data by merging the probabilistic reasoning of Bayesian networks with the interpretability of decision trees, the pattern recognition abilities of ANNs, and the temporal interdependence of Markov models. This helps doctors to make more educated judgments based on a larger spectrum of information, leading to better patient outcomes. Healthcare workers can get to vital data from any place because to the cloud-enabled architecture's seamless scalability and accessibility. This not only increases the efficiency of decision-making, but also improves communication and cooperation between different medical professionals. This uses cutting-edge modeling strategies and cloud computing to pave a new path in clinical decision-making. This system has the potential to greatly enhance healthcare by integrating image-driven insights with CDSS, to the advantage of both patients and healthcare practitioners. [ABSTRACT FROM AUTHOR]

With the increasing global population and escalating ecological and farmland degradation, challenges to the environment and livelihoods have become prominent. Coordinating urban development, food security, and ecological conservation is crucial for fostering sustainable development. This study focuses on assessing the "Ecology-Agriculture-Urban" (E-A-U) space in Yulin City, China, as a representative case. Following the framework proposed by Chinese named "environmental capacity and national space development suitability evaluation" (hereinafter referred to as "Double Evaluation"), we developed a Self-Attention Residual Neural Network (SARes-NET) model to assess the E-U-A space. Spatially, the northwest region is dominated by agriculture, while the southeast is characterized by urban and ecological areas, aligning with regional development patterns. Comparative validations with five other models, including Logistic Regression (LR), Naive Bayes (NB), Gradient Boosting Decision Trees (GBDT), Random Forest (RF) and Artificial Neural Network (ANN), reveal that the SARes-NET model exhibits superior simulation performance, highlighting it's ability to capture intricate non-linear relationships and reduce human errors in data processing. This study establishes deep learning-guided E-A-U spatial evaluation as an innovative approach for national spatial planning, holding broader implications for national-level territorial assessments. [ABSTRACT FROM AUTHOR]

Introduction: Additive manufacturing (AM) is a revolutionary technology transforming traditional production processes by providing exceptional mechanical characteristics. Methods: The present study aims explicitly to predict the hardness of Polycarbonate (PC) parts produced using AM. The objectives of this study are: (1) To investigate the process parameters that impact the ability to estimate the hardness of PC materials accurately, and (2) To develop a best-performing ML model from a range of models that can reliably predict the hardness of additively manufactured PC parts. Initially, fused filament fabrication (FFF), the most affordable AM technique, was used for the manufacturing of parts. Four process parameters, infill density, print direction, raster angle, and layer thickness, are selected for investigation. A heatmap is generated to obtain the influence of process parameters on hardness. Then, machine learning (ML) techniques create a range of predictive models that can predict hardness value considering the level of process parameters. Results: The developed ML models include Linear Regression, Decision Tree, Random Forest, K-nearest neighbor, Support Vector Regression, AdaBoost, and Artificial Neural Network. Further, an investigation has been done that includes choosing and improving ML algorithms and assessing the models' performance. Discussion: Prediction plots, residual plots, and evaluation metrics plots are prepared to gauge the performance of the developed models. Thus, the research enhances AM capabilities by applying predictive modeling to process parameters and improving the quality and reliability of fabricated components. [ABSTRACT FROM AUTHOR]

Construction firms face considerable challenges in relation to finding cost-effective formwork solutions to meet increased construction demands. Project stakeholders have relied on self-consolidating concrete (SCC) to speed up the construction time because SCC is highly fluid and has numerous advantages compared to traditional concrete. To withstand SCC's high fluidity, formwork systems should be robust. Although previous research has experimentally examined various characteristics of SCC, few research studies have used machine-learning algorithms to estimate or predict the lateral pressure exerted by SCC on formwork systems. Hence, this study addressed this knowledge gap by proposing a machine-learning approach to predict the lateral pressure of SCC on vertical formwork systems. First, laboratory tests were performed to collect data on lateral pressure measurements, material factors, placement conditions, and formwork characteristics affecting the SCC lateral pressure on formwork systems. Second, four supervised machine-learning algorithms were considered in this study: k-nearest neighbor (KNN), artificial neural network (ANN), decision tree (DT), and random forest (RF). Third, the hyperparameters of the machine-learning algorithms were tuned, and their performance metrics were compared. Fourth, the most accurate predictive machine-learning model was verified on an unseen testing set. The results showed that the RF machine-learning algorithm was the best model for predicting the lateral pressure of SCC on formwork systems, with a mean percentage error of 0.8%, a mean absolute percentage error of 4.29%, and a coefficient of determination R2 of 0.9548. This study adds to the construction engineering and management body of knowledge by developing a machine-learning predictive model that can be used to accurately assess the lateral pressure exerted by SCC on formwork, which helps to ensure safe design of formwork systems and economic construction operations in formwork-related activities. [ABSTRACT FROM AUTHOR]

This research primary aimed at evaluating various predictive models in predicting programming students at risk of dropping out. It also aimed at identifying attributes that are significant in predicting students at risk of dropping. The educational data mining process (EDM) was utilized as the research framework. The study conducted a ten-fold cross-validation, revealing that the k-nearest neighbors (kNN) algorithm achieved the highest classification accuracy at 95.5%. The decision tree model followed closely with a 94.9% accuracy, logistic regression exhibited 94.4%, and the neural network model yielded a classification accuracy of 93.2%. Further analysis, including confusion matrices and receiver operating characteristic (ROC) curves, provided detailed insights into the models' performance. Notably, the decision tree algorithm excelled in identifying students who did not drop out, with a misclassification rate of 9 out of 30 for dropped students. Analysis also showed that students' assignments completed (AC), laboratory work (LW), and attendance (ATT) were the strongest predictors in identifying students at risk of dropping. Results of the study can be used by instructors to identify in advance student at risk of dropping and provide them with the necessary intervention to improve performance in programming. [ABSTRACT FROM AUTHOR]

One expanding area of bioinformatics is medical diagnosis through the categorization of biomedical characteristics. Automatic medical strategies to boost the diagnostic through machine learning (ML) methods are challenging. They require a formal examination of their performance to identify the best conditions that enhance the ML method. This work proposes variants of the Voting and Stacking (VC and SC) ensemble strategies based on diverse auto-tuning supervised machine learning techniques to increase the efficacy of traditional baseline classifiers for the automatic diagnosis of vertebral column orthopedic illnesses. The ensemble strategies are created by first combining a complete set of auto-tuned baseline classifiers based on different processes, such as geometric, probabilistic, logic, and optimization. Next, the three most promising classifiers are selected among k-Nearest Neighbors (kNN), Naïve Bayes (NB), Logistic Regression (LR), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machine (SVM), Artificial Neural Networks (ANN), and Decision Tree (DT). The grid-search K-Fold cross-validation strategy is applied to auto-tune the baseline classifier hyperparameters. The performances of the proposed ensemble strategies are independently compared with the auto-tuned baseline classifiers. A concise analysis evaluates accuracy, precision, recall, F1-score, and ROC-ACU metrics. The analysis also examines the misclassified disease elements to find the most and least reliable classifiers for this specific medical problem. The results show that the VC ensemble strategy provides an improvement comparable to that of the best baseline classifier (the kNN). Meanwhile, when all baseline classifiers are included in the SC ensemble, this strategy surpasses 95% in all the evaluated metrics, standing out as the most suitable option for classifying vertebral column diseases. [ABSTRACT FROM AUTHOR]

A modified multi-layer perceptron (MLP) model based on decision trees (DT-MLP) is presented to predict velocity and water free-surface profiles in a 90° open-channel bend. The ability of the new hybrid model to predict the velocity and flow depth in a 90° sharp bend is investigated and compared with the abilities of MLP and multiple-linear regression (MLR) models. The MLP and DT-MLP networks are trained and tested using 520 and 506 experimental data measured for velocity and flow depth, respectively, at five different discharge rates of 5, 7.8, 13.6, 19.1 and 25.3 l/s. The MLP and DT-MLP comparison results against MLR reveal that the two artificial neural networks (ANNs) are 84% and 16% more accurate than the MLR model in predicting the velocity and flow depth variables, respectively. According to the results, the root mean square error (RMSE) value of the DT-MLP model decreases by 9% and 7.5% in predicting velocity and flow depth, respectively, compared with the MLP model. It was found that the hybrid decision-tree-based method can significantly improve MLP neural network performance in forecasting velocity and free-surface profiles in a 90° open-channel bend. [ABSTRACT FROM AUTHOR]

The paper explores the possibilities of using data classification methods when forecasting time series of the geomagnetic Kp-index by machine learning methods. To classify categories of the Kp-index based on the degree of disturbance, linear and logistic regression, random forest, gradient boosting on top of decision trees, and artificial neural networks of various architectures are used. The results of these methods are compared with a trivial inertial forecast (the statistical indicators of which for problems of this type are always high) at horizons from 3 h to 1 day in 3-h increments. The problem of choosing a cross-validation scheme for selecting the model hyperparameters, ways to overcome the imbalance of categories, the relative importance of input features, as well as the dependence of the results on the test sample (beginning of the 25th solar activity cycle) on inclusion in the training sample of data from the 23rd and 24th cycles or only the 24th cycles are studied. Based on the results, conclusions are drawn about the preferred methods for classifying values of the Kp-index based on the level of geomagnetic disturbance. Ways for further research and possible improvement of the classification quality are outlined, including for determining the characteristic hidden states of Earth's magnetosphere as a dynamic system in order to improve the quality of forecasting geomagnetic indices. [ABSTRACT FROM AUTHOR]

Copyright of Fire Safety Science is the property of Fire Safety Science Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

BACKGROUND: Numerous systems for detecting driver drowsiness have been developed; however, these systems have not yet been widely used in real-time. OBJECTIVE: The purpose of this study was to investigate at the feasibility of detecting alert and drowsy states in drivers using an integration of features from respiratory signals, vehicle lateral position, and reaction time and out-of-vehicle ways of data collection in order to improve the system's performance and applicability in the real world. METHODS: Data was collected from 25 healthy volunteers in a driving simulator-based study. Their respiratory activity was recorded using a wearable belt and their reaction time and vehicle lateral position were measured using tests developed on the driving simulator. To induce drowsiness, a monotonous driving environment was used. Different time domain features have been extracted from respiratory signals and combined with the reaction time and lateral position of the vehicle for modeling. The observer of rating drowsiness (ORD) scale was used to label the driver's actual states. The t-tests and Man-Whitney test was used to select only statistically significant features (p < 0.05), that can differentiate between the alert and drowsy states effectively. Significant features then combined to investigate the improvement in performance using the Multilayer Perceptron (MLP), the Support Vector Machines (SVMs), the Decision Trees (DTs), and the Long Short Term Memory (LSTM) classifiers. The models were implemented in Python library 3.6. RESULTS: The experimental results illustrate that the support vector machine classifier achieved accuracy of 88%, precision of 85%, recall of 83%, and F1 score of 84% using selected features. CONCLUSION: These results indicate the possibility of very accurate detection of driver drowsiness and a viable solution for a practical driver drowsiness system based on combined measurement using less-intrusive and out-of-vehicle recording methods. [ABSTRACT FROM AUTHOR]

Research background: Financial distress or non-prosperity prediction has been a widely discussed topic for several decades. Early detection of impending financial problems of the company is crucial for effective risk management and important for all entities involved in the company's business activities. In this way, it is possible to take the actions in the management of the company and eliminate possible undesirable consequences of these problems. Purpose of the article: This article aims to innovate financial distress prediction through the creation of individual models and ensembles, combining machine learning techniques such as decision trees and neural networks. These models are developed using real data. Beyond serving as an autonomous and universal tool especially useful in the Slovak economic conditions, these models can also represent a benchmark for Central European economies confronting similar economic dynamics. Methods: The prediction models are created using a dataset consisting of more than 20 financial ratios of more than 19 thousand real companies. Partial models are created employing machine learning algorithms, namely decision trees and neural networks. Finally, all models are compared based on a wide range of selected performance metrics. During this process, we strictly use a data mining methodology CRISP-DM. Findings & value added: The research contributes to the evolution of financial prediction and reveals the effectiveness of ensemble modelling in predicting financial distress, achieving an overall predictive ability of nearly 90 percent. Beyond its Slovak origins, this study provides a framework for early financial distress prediction. Although the models are created for diverse industries within the Slovak economy, they could also be useful beyond national borders. Moreover, the CRISP-DM methodological framework enables its adaptability for companies in other countries. [ABSTRACT FROM AUTHOR]

When designing helicon plasma thrusters, one important characteristic is the impedance of the radio-frequency antenna that is used to deposit power into the plasma. This impedance can be characterized both experimentally and numerically. Recently, a numerical tool capable of predicting the antenna impedance, called Adamant, has been developed. However, Adamant takes a long time to run and has high computer resource demands. Therefore, this work has been done to evaluate whether machine learning models, trained on Adamant-generated data, can be used instead of Adamant for small design change evaluations and similar works. Six different machine learning models were implemented in MATLAB: decision trees, ensembles, support vector machines, Gaussian process regressions, generalized additive models and artificial neural networks. These were trained and evaluated using nested k-fold cross-validation with the hyperparameters selected using Bayesian optimization. The performance target was to have less than 5% error on a point-to-point basis. The artificial neural network performed the best when taking into account both maximum error magnitudes and generalization ability, with a maximum error of 3.98% on the test set and with considerably better performance than the other models when tested on some practical examples. Future work should look into different solver algorithms for the artificial neural network to see if the results could be improved even further. To expand the model's usefulness it might also be worth looking into implementing different antenna types that are of interest for helicon plasma thrusters. [ABSTRACT FROM AUTHOR]

Given the climate change in recent decades and the ever-increasing energy consumption in the building sector, research is widely focused on the green revolution and ecological transition of buildings. In this regard, artificial intelligence can be a precious tool to simulate and optimize building energy performance, as shown by a plethora of recent studies. Accordingly, this paper provides a review of more than 70 articles from recent years, i.e., mostly from 2018 to 2023, about the applications of machine/deep learning (ML/DL) in forecasting the energy performance of buildings and their simulation/control/optimization. This review was conducted using the SCOPUS database with the keywords "buildings", "energy", "machine learning" and "deep learning" and by selecting recent papers addressing the following applications: energy design/retrofit optimization, prediction, control/management of heating/cooling systems and of renewable source systems, and/or fault detection. Notably, this paper discusses the main differences between ML and DL techniques, showing examples of their use in building energy simulation/control/optimization. The main aim is to group the most frequent ML/DL techniques used in the field of building energy performance, highlighting the potentiality and limitations of each one, both fundamental aspects for future studies. The ML approaches considered are decision trees/random forest, naive Bayes, support vector machines, the Kriging method and artificial neural networks. The DL techniques investigated are convolutional and recursive neural networks, long short-term memory and gated recurrent units. Firstly, various ML/DL techniques are explained and divided based on their methodology. Secondly, grouping by the aforementioned applications occurs. It emerges that ML is mostly used in energy efficiency issues while DL in the management of renewable source systems. [ABSTRACT FROM AUTHOR]

Simple Summary: In this work, we explored computational methods for analyzing a color digital image of a wound and predicting (from the analyzed image) the number of days it will take for the wound to fully heal. We used a hybrid computational approach combining deep neural networks and decision trees, and within this hybrid approach, we explored (and compared the accuracies of) different types of models for predicting the time to heal. More specifically, we explored different models for finding the outline of the wound within the wound image and we proposed a model for computing the proportions of different types of tissues within the wound bed (e.g., fibrin slough, granulation, or necrotic tissue). Our work clarifies what type of model should be used for the computational prediction of wound time-to-healing and establishes that, in order to predict time-to-healing accurately, it is important to incorporate (into the model) data on the proportions of different types in the wound bed. Despite the societal burden of chronic wounds and despite advances in image processing, automated image-based prediction of wound prognosis is not yet in routine clinical practice. While specific tissue types are known to be positive or negative prognostic indicators, image-based wound healing prediction systems that have been demonstrated to date do not (1) use information about the proportions of tissue types within the wound and (2) predict time-to-healing (most predict categorical clinical labels). In this work, we analyzed a unique dataset of time-series images of healing wounds from a controlled study in dogs, as well as human wound images that are annotated for the tissue type composition. In the context of a hybrid-learning approach (neural network segmentation and decision tree regression) for the image-based prediction of time-to-healing, we tested whether explicitly incorporating tissue type-derived features into the model would improve the accuracy for time-to-healing prediction versus not including such features. We tested four deep convolutional encoder–decoder neural network models for wound image segmentation and identified, in the context of both original wound images and an augmented wound image-set, that a SegNet-type network trained on an augmented image set has best segmentation performance. Furthermore, using three different regression algorithms, we evaluated models for predicting wound time-to-healing using features extracted from the four best-performing segmentation models. We found that XGBoost regression using features that are (i) extracted from a SegNet-type network and (ii) reduced using principal components analysis performed the best for time-to-healing prediction. We demonstrated that a neural network model can classify the regions of a wound image as one of four tissue types, and demonstrated that adding features derived from the superpixel classifier improves the performance for healing-time prediction. [ABSTRACT FROM AUTHOR]

Predicting bankruptcy within selected industries is crucial because of the potential ripple effects and unique characteristics of those industries. It serves as a risk management tool, guiding various stakeholders in making decisions. While artificial intelligence (AI) has shown high success rates in classification tasks, it remains uncertain whether its use significantly enhances the potential for early warning of impending problems. The following question arises: will classical methods eventually replace the effectiveness of these advanced techniques? This paper sheds light on the fact that even classical methods continue to achieve results that are not far behind, highlighting their enduring importance in financial analysis. This paper aims to develop bankruptcy prediction models for the chemical industry in Slovakia and to compare their effectiveness. Predictions are generated using the classical logistic regression (LR) method as well as AI techniques, artificial neural networks (ANNs), support vector machines (SVMs), and decision trees (DTs). The analysis aims to determine which of the employed methods is the most efficient. The research sample consists of circa 600 enterprises operating in the Slovak chemical industry. The selection of eleven financial indicators used for bankruptcy prediction was grounded in prior research and existing literature. The results show that all of the explored methods yielded highly similar outcomes. Therefore, determining the clear superiority of any single method is a difficult task. This might be partially due to the potentially reduced quality of the input data. In addition to classical statistical methods employed in econometrics, there is an ongoing development of AI-based models and their hybrid forms. The following question arises: to what extent can these newer approaches enhance accuracy and effectiveness? [ABSTRACT FROM AUTHOR]

Seawalls are critical defence infrastructures in coastal zones that protect hinterland areas from storm surges, wave overtopping and soil erosion hazards. Scouring at the toe of sea defences, caused by wave-induced accretion and erosion of bed material imposes a significant threat to the structural integrity of coastal infrastructures. Accurate prediction of scour depths is essential for appropriate and efficient design and maintenance of coastal structures, which serve to mitigate risks of structural failure through toe scouring. However, limited guidance and predictive tools are available for estimating toe scouring at sloping structures. In recent years, Artificial Intelligence and Machine Learning (ML) algorithms have gained interest, and although they underpin robust predictive models for many coastal engineering applications, such models have yet to be applied to scour prediction. Here we develop and present ML-based models for predicting toe scour depths at sloping seawall. Four ML algorithms, namely, Random Forest (RF), Gradient Boosted Decision Trees (GBDT), Artificial Neural Networks (ANNs), and Support Vector Machine Regression (SVMR) are utilised. Comprehensive physical modelling measurement data is utilised to develop and validate the predictive models. A Novel framework for feature selection, feature importance, and hyperparameter tuning algorithms are adopted for pre- and post-processing steps of ML-based models. In-depth statistical analyses are proposed to evaluate the predictive performance of the proposed models. The results indicate a minimum of 80% prediction accuracy across all the algorithms tested in this study and overall, the SVMR produced the most accurate predictions with a Coefficient of Determination (r2) of 0.74 and a Mean Absolute Error (MAE) value of 0.17. The SVMR algorithm also offered most computationally efficient performance among the algorithms tested. The methodological framework proposed in this study can be applied to scouring datasets for rapid assessment of scour at coastal defence structures, facilitating model-informed decision-making. [ABSTRACT FROM AUTHOR]

The number of seniors living alone without caretakers is on the rise. With falls accounting for a significant percentage of injuries and mortality in the elderly, fall detection devices are critical, allowing immediate medical attention through promptly alerting caretakers. We reviewed current fall detection methods and improved their accuracy by implementing new considerations. A "double-checking" system minimizes errors by combining vision and vibration-based detection reports using decision trees. Adjustments to existing vision-based detection methods were made by improving camera positioning and introducing angular velocity. We reviewed various vision and vibration-based fall detection models, analyzed their inconsistency patterns, and optimally consolidated their results. [ABSTRACT FROM AUTHOR]

Copyright of Transformations in Business & Economics is the property of Vilnius University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Air pollution is a global issue of concern. Accurate prediction of the air quality index (AQI) plays a crucial role in understanding pollution patterns and implementing effective mitigation strategies. This research paper presents a comprehensive study on the application of ensemble learning techniques for AQI prediction, aiming to enhance prediction accuracy and provide insights into the driving factors behind pollution levels. The primary objective is to conduct a comparative analysis of homogenous ensemble method, namely, bagging in conjunction with individual base models, such as multiple linear regression, support vector regression, decision trees, and artificial neural networks. In total we have created 4 non-ensembled models and 4 ensembled models. Additionally, the research investigates the importance of different features in AQI prediction models, aiming to identify the key factors influencing air pollution. Hyperparameter tuning is done to optimize the models. [ABSTRACT FROM AUTHOR]

Objectives: This study aims to introduce a prediction model based on a machine learning approach as an efficient solution for prediction purposes to better prognosis and increase CRC survival. Methods: In the current retrospective study, we used the data of 1062 CRC cases to analyse and establish a prediction model for the 5-year CRC survival. The machine learning algorithms were used to develop prediction models, including random Forest, XG-Boost, bagging, logistic regression, support vector machine, artificial neural network, decision tree, and K-nearest neighbours. Results: The current study revealed that the XG-Boost with AU-ROC of 0.906 and 0.813 for internal and external conditions gave us better insight into predictability and generalizability than other algorithms. Conclusion: XG-Boost can be utilised as a knowledge source for implementing intelligent systems as an assistive tool for clinical decision-making in healthcare settings to improve prognosis and increase CRC survival through various clinical solutions that doctors can achieve. [ABSTRACT FROM AUTHOR]

This study evaluated the performance of multiple models that used machine learning to anticipate wind speed in the city of Dhaka. The NASA Power website provided the data set for this investigation. The models used for prediction included the decision tree regressor, support vector regressor, random forest, linear regression, neural network and polynomial regression. A hold-out check and k-fold cross-validation were used to assess how well these models performed. With the highest R2 scores and lowest RMSEs on both the validation and test sets, the results demonstrated that the polynomial regression model performed the best. With the lowest R2 scores and largest RMSEs on both sets, the decision tree model scored the poorest. High R2 scores and low RMSEs were achieved by the random forest model, which had comparable performance to the polynomial regression model but required a longer computation time. In addition, the neural network model demonstrated commendable predictive accuracy, yielding an R2 score of 0.67 and a low RMSE of 0.57. However, its application is contingent on the availability of substantial computational resources, given its extensive computation time of 457.93 s. The study concludes by highlighting the efficacy of the Polynomial Regression model as the optimal choice for wind speed prediction in Dhaka, offering a balance between superior performance and efficient computation. This insight provides valuable guidance for practitioners and researchers seeking effective models for similar applications. [ABSTRACT FROM AUTHOR]

The process of categorizing students' performance based on input data, encompassing demographic information and final exam results, is recognized as student performance classification. Educational data mining has gained traction in assessing students' performance. However, this study entails the need to analyze the diverse attributes of students' information within an educational institution by using data mining techniques. This study thoroughly examines both previous and current methodologies presented by researchers, addressing two main aspects: data preprocessing and classification algorithms applied in student performance classification. Data preprocessing specifically delves into the exploration of feature selection techniques, encompassing three types of feature selection and search methods. These techniques aim to identify the most significant features, eliminate unnecessary ones, and reduce data dimensionality. In addition, classification algorithms play a crucial role in categorizing or predicting student performance. Models such as k-nearest neighbors (KNN), decision tree (DT), artificial neural networks (ANN), and linear models (LR) were scrutinized based on their performance in prior research. Ultimately, this study highlights the potential for further exploration of feature selection techniques like information gain, Chi-square, and sequential selection, particularly when applied to new datasets such as students' online learning activities, utilizing a variety of classification algorithms. [ABSTRACT FROM AUTHOR]

Breast cancer represents a significant health concern, particularly in Saudi Arabia, where it ranks as the most prevalent cancer type among women. This study focuses on leveraging eXplainable Artificial Intelligence (XAI) techniques to predict benign and malignant breast cancer cases using various clinical and pathological features specific to Saudi Arabian patients. Six distinct models were trained and evaluated based on common performance metrics such as accuracy, precision, recall, F1 score, and AUC-ROC score. To enhance interpretability, Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) were applied. The analysis identified the Random Forest model as the top performer, achieving an accuracy of 0.72, along with robust precision, recall, F1 score, and AUC-ROC score values. Conversely, the Support Vector Machine model exhibited the poorest performance metrics, indicating its limited predictive capability. Notably, the XAI approaches unveiled variations in the feature importance rankings across models, underscoring the need for further investigation. These findings offer valuable insights into breast cancer diagnosis and machine learning interpretation, aiding healthcare providers in understanding and potentially integrating such technologies into clinical practices. [ABSTRACT FROM AUTHOR]

Power quality improvement and Power quality disturbance (PQD) detection are two significant concerns that must be addressed to ensure an efficient power distribution within the utility grid. When the process to analyze PQD is migrated to real-time platforms, the possible occurrence of a phase mismatch can affect the algorithm's accuracy; this paper evaluates phase shifting as an additional stage in signal acquisition for detecting and classifying eight types of single power quality disturbances. According to their mathematical models, a set of disturbances was generated using an arbitrary waveform generator BK Precision 4064. The acquisition, detection, and classification stages were embedded into a BeagleBone Black. The detection stage was performed using multiresolution analysis. The feature vectors of the acquired signals were obtained from the combination of Shannon entropy and log-energy entropy. For classification purposes, four types of classifiers were trained: multilayer perceptron, K-nearest neighbors, probabilistic neural network, and decision tree. The results show that incorporating a phase-shifting stage as a preprocessing stage significantly improves the classification accuracy in all cases. [ABSTRACT FROM AUTHOR]

Seismic liquefaction has been reported in sandy soils as well as gravelly soils. Despite sandy soils, a comprehensive case history record is still lacking for developing empirical, semi-empirical, and soft computing models to predict this phenomenon in gravelly soils. This work compiles documentation from 234 case histories of gravelly soil liquefaction from across the world to generate a database, which will then be used to develop seismic gravelly soil liquefaction potential models. The performance measures, namely, accuracy, precision, recall, F-score, and area under the receiver operating characteristic curve, were used to evaluate the training and testing tree-based models' performance and highlight the capability of the logistic model tree over reduced error pruning tree, random tree and random forest models. The findings of this research can provide theoretical support for researchers in selecting appropriate tree-based models and improving the predictive performance of seismic gravelly soil liquefaction potential. [ABSTRACT FROM AUTHOR]

In addressing the prevalent challenges associated with the limited generalization ability and weak adaptability to small samples observed in neural network models in contemporary yarn quality prediction research, this study proposes a two-layer stacking regression model. The initial layer of the stacking model integrates three base models: random forest, gradient boosting decision tree, and adaptive boosting. The secondary layer employs the linear regression algorithm. The efficacy of the stacking model was juxtaposed against the performance assessment of a multilayer perceptron neural network model, in addition to evaluations of the single random forest, gradient boosting decision tree, and adaptive boosting models, respectively. The generalization ability of the models was assessed through experiments that randomly partitioned the sample dataset based on 10 distinct random seeds. The adaptability to small samples of the models was evaluated through experiments spanning nine varying training set sample sizes. The experiment results underscore that the stacking model outperforms in aspects of prediction accuracy, generalization ability, and adaptability to small samples. [ABSTRACT FROM AUTHOR]

Carbonation is one of the critical issues affecting the durability of reinforced concrete. Evaluating the depth of concrete carbonation is of great significance for ensuring the quality and safety of construction projects. In recent years, various prediction algorithms have been developed for evaluating concrete carbonation depth. This article provides a detailed overview of the existing prediction models for concrete carbonation depth. According to the data processing methods used in the model, the existing prediction models can be divided into mathematical curve models and machine learning models. The machine learning models can be further divided into the following categories: artificial neural network model, decision tree model, support vector machine model, and combined models. The basic idea of the mathematical curve model is to directly establish the relationship between the carbonation depth and age of concrete by using certain function curves. The advantage of the mathematical curve model is that only a small amount of experimental data is needed for curve fitting, which is very convenient for engineering applications. The limitation of the curve model is that it can only consider the influence of some factors on the carbonation depth of concrete, and the prediction accuracy cannot be guaranteed. The advantage of using the machine learning model to predict the carbonation depth of concrete is that many factors can be considered at the same time. When there are sufficient experimental data, the trained machine learning model can give more accurate prediction results than the mathematical curve model. The main defect of the machine learning model is that it needs a lot of experimental data as training samples, so it is not as convenient as the mathematical curve model in engineering applications. A future research direction may be to combine a machine learning model with a mathematical curve model to evaluate the carbonation depth of concrete more accurately. [ABSTRACT FROM AUTHOR]

When a public transport manager conducts a customer satisfaction survey (CSS), the goal is to determine the overall satisfaction of passengers with the service, as well as their satisfaction with specific aspects (e.g. frequency, speed, and comfort). Another fundamental objective is to assess the importance to customers of each attribute individually. Asking directly about this importance involves a number of drawbacks; therefore, most studies extract this importance from surveys that ask questions only about global satisfaction and specific satisfaction regarding each attribute. This paper investigates the capability and performance of two emerging data mining methods, namely, decision trees and neural networks, for extracting the importance of attributes from CSS. A total of 858 surveys about the metropolitan bus service in Granada (Spain) were used to model estimation and evaluation. The main advantages and disadvantages of each method are studied from the standpoint of public transport managers. [ABSTRACT FROM AUTHOR]

When descriptions of data values are too detailed, the computational complexities involved in mining useful knowledge from the database generally increases. This gives rise to the need of tools techniques which can reduce these complexities and mine the valuable information hidden behind the database. There exists number of such techniques viz. decision trees, neural networks, rough-set theory, rule induction, and case-based reasoning which are able to meet the aforesaid objective up to some extent. Each of these techniques has its advantages and limitations that motivate researchers to develop new tools for the mining tasks. In this paper, we have developed a novel methodology, genetically optimized cluster oriented soft decision trees (GCSDT), to glean vital information imbedded in the large databases. In contrast to the standard C-fuzzy decision trees, where granules are developed through fuzzy (soft) clustering, in the proposed architecture granules are developed by means of genetically optimized soft clustering. In the GCSDT architecture, GA ameliorates the difficulty of choosing an initialization for the fuzzy clustering algorithm and always avoids degenerate partitions. This provides an effective means for the optimization of clustering criterion, where an objective function can be illustrated in terms of cluster’s center. Growth of the GCSDT is realized by expanding nodes of the tree, characterized by the highest inconsistency index of the information granules. In order to validate the proposed tree structure it has been deployed on synthetic and machine learning data sets. Moreover, Results are compared with those produced by standard C4.5 decision trees and C-fuzzy decision trees; further student t-test is applied to show that these differences in results are statistically significant. [Copyright &y& Elsevier]