Search

Non-linear weirs meanwhile economic advantages, have more passing flow capacity than linear weirs. These weirs have higher discharge efficiency with less free height upstream compared to linear weirs by increasing the length of the crown at a certain width. Intelligent algorithms have found a valuable place among researchers due to their great ability to discover complex and hidden relationships between effective independent parameters and dependent parameters, as well as saving money and time. In this research, the performance of support vector machine (SVM) and gene expression programming algorithm (GEP) in predicting the discharge coefficient of arched non-linear weirs was investigated using 243 laboratory data series for the first scenario and 247 laboratory data series for the second scenario. The geometric and hydraulic parameters were used in this research including the water load (HT), weir height (P), total water load ratio (HT/p), arc cycle angle (Ɵ), cycle wall angle (α), and discharge coefficient (Cd). The results of artificial intelligence showed that the combination of parameters (Cd, H_T/p, α, Ɵ) respectively in GEP and SVM algorithms in the training phase related to the first scenario (Labyrinth weir with cycle wall angle 6 degrees) were respectively equal to (R²=0.9811), (RMSE=0.02120), (DC=0.9807), and (R²=0.9896), (RMSE=0.0189), (DC=0.9871) in the second scenario (Labyrinth weir with a cycle wall angle of 12 degrees) it was equal to (R²=0.9770), (RMSE=0.0193), (RMSE=0.9768), and (R² = 0.9908), (RMSE = 0.0128), (DC = 0.9905), which compared to other combinations has led to the most optimal output that shows the very favorable accuracy of both algorithms in predicting the coefficient the Weir discharge is arched non-linear. The results of the sensitivity analysis indicated that the effective parameter in determining the discharge coefficient of the arched non-linear Weir in GEP and in SVM is the total water load ratio parameter (HT/p). Comparing the results of this research with other researchers revealed that the evaluation indices for GEP and SVM algorithms of this research had better estimates than other researchers. [ABSTRACT FROM AUTHOR]

Using transfer functions to predict soil moisture parameters has been considered strictly a scientific and economical method among researchers. In this research, field capacity (FC) and permanent wilting point (PWP) of soil were predicted using classic regression (linear and non-linear), support vector machine (SVM) algorithm, and gene programming expression (GEP) algorithm based on three performance assessment criteria as determination of coefficient (R² ), root mean square error (RMSE), and standardized developed discrepancy Ratio (DDR) in the Arasbaran plain in the northwest of Iran. Independent parameters were determined as clay percent (Cl), silt percent (Si), gravel percent (Sa), organic carbon (OC), bulk density (ρb), and actual density (ρs) which (S, ρb, ρs) and (ρb, ρs) were opted to predict FC and PWP using Gamma test, respectively. The results showed that each three transfer functions are capable to simulate FC and PWP parameters but the SVM algorithm is the superior predictor among the three functions so the values of (R², RMSE, and DDRmax) of training and testing phases for FC were obtained (0.9908, 0.5517, 17.50), (0.9785, 0.7004, 11.62) and those of PWP were calculated (0.9782, 0.5764, 2.85) and (0.8389, 1.187, 3.09), respectively. [ABSTRACT FROM AUTHOR]

Background & Aims: According to the report released by world health organization (WHO), the ST-segment elevation myocardial infarction- cardiogenic shock (STEMI-CS) is one of the important factors in patient mortality within hospitals (1), (2), (3), (4). CS and its related complications need a huge financial and medical burden. Some researchers stated that high mortality and complication rates of STEMI-CS patients are associated with the lack of effective early preventive treatments. Given the risk of CS and the different risk factors associated with it, accurate clinical risk prediction tools need to be developed to accurately predict the onset of CS. Recently, researchers have been used various machine learning methods to predict the risk of mortality in STEMI-CS patients. Recently, machine learning (ML) methods were developed to establish predictive models to identify the in-hospital mortality risk of STEMI-CS patients. The existing methods achieved encouraging results; however, their performance is not ideal, and more effort is needed to improve the performance. The aim of this study is to present a hybrid machine learning method for predicting the risk of mortality in STEMI-CS patients. Our proposed method combines a powerful swarm intelligence strategy, anti-coronavirus optimization algorithm (ACVO) with support vector machine (SVM) in risk prediction phase. The proposed model is compared with standard support vector machine (SVM), least absolute shrinkage and selection operator (LASSO), and adaptive neuro fuzzy inference system (ANFIS) on a real-world benchmark dataset. Methods: To predict the mortality status of STEMI-CS patients, we proposed the ACVOSVM algorithm. The proposed method is a hybrid machine learning algorithm that combines the SVM with ACVO algorithm to identify the most effective parameters on the death of patients. The incentive mechanism of using ACVO is to optimally configure the parameters of SVM to improve its prediction performance. The proposed ACO-SVM is also useful in determining the optimal subset of features and treatment strategies that have the greatest impact in predicting the status of STEMI-CS patients. The proposed approach models the problem of predicting the status of patients as an optimization problem. In order to determine the most effective features in predicting the survival or death of STEMI-CS patients, the proposed ACVO-SVM model is trained with different combinations of patient characteristics and adopted treatment strategies. Then the best combination of features that provides the highest performance is considered as the superior combination. To select the most effective features, first all the features are considered for training the SVM model, then the remaining features are ignored one by one and the model with the same structure is trained. The models were compared based on accuracy, recall rate, F1 criterion. Finally, the best model is used to predict the status of patients in test dataset. The data set used to evaluate the proposed method includes 410 records of patients hospitalized due to STEMI-CS complications in Shahid Madani Hospital of Tabriz University of Medical Sciences. The collected data is related to a 10-year period from 2009 to 2018. This data set includes five categories of main characteristics, which are demographic characteristics, type of myocardial infarction, risk factors, clinical symptoms, and type of treatment used. It should be noted that 80% of the records of the data set are considered as training data, and 20% of the records are considered as the test data set. The proposed method is implemented in MATLAB software. Results: Among M1 to M5 feature combination models, the experimental results show that the M1 model has higher performance on the training and test dataset in terms of predicting the patient's condition compared to other combination models. Model M1 includes the combination of characteristics of age, sex, type of myocardial infarction, smoking, percutaneous vascular interventions and coronary artery bypass surgery. This shows that considering the mentioned features has the greatest effect on the final condition of STEMI-CS patients. The results are in line with previous studies (2), (3) in this field, which stated that age, gender, smoking, coronary artery bypass surgery and percutaneous vascular interventions have the greatest effect on the mortality rate of patients. The M2 model ranks second in terms of efficiency in determining the status of patients, which shows that smoking also has a greater effect on the mortality of patients with STEMI-CS. Also, the M3 model indicates that the use of the balloon pump treatment strategy, along with other demographic symptoms of the patient, history of heart infarction and smoking have a great effect on the mortality rate of patients. In summary, it can be concluded that the demographic characteristics of the patient such as age and gender, smoking, history of illness and the use of coronary bypass surgery and percutaneous vascular interventions have a great impact on the mortality of STEMI-CS patients. The proposed ACVO-SVM approach is compared with several other popular approaches, which include: standard SVM model, LASSO regression, ANFIS, and XGBoost. The experimental results justify that the proposed ACVO-SVM outperformed its counterparts. Conclusion: In this study, a hybrid supervised machine learning model was presented to determine the status of patients with cardiogenic shock due to ST-segment elevation myocardial infarction. The proposed ACVO-SVM model uses an ACVO optimization algorithm to estimate the optimal parameters of the SVM model, making the SVM training process more efficient. The proposed model was evaluated using a dataset of patients with cardiogenic shock and the results were compared with the LASSO, ANFIS, XGBoost, and SVM models. The results showed that the proposed method worked well compared to other proposed classification models. We also found that age, gender, type of myocardial infarction, smoking, percutaneous vascular surgery, and coronary bypass transplantation surgery are the most effective factors for survival in STEMI-CS patients. In this research, the models were evaluated on a small dataset. Therefore, one of the necessary tasks to improve this research is to evaluate the proposed method and other counterpart models on large datasets to determine their strengths and weaknesses. Another limitation of this research is the lack of examination of all factors affecting the survival of STEMI-CS patients, such as blood sugar level and duration of ischemia. For this reason, it is necessary to investigate all factors affecting the mortality of STEMI-CS patients to improve the quality of classification and prediction of the final status of patients. [ABSTRACT FROM AUTHOR]

In the present study, the flow rate in flues containing lateral semi-cylinders (SMBF) was simulated and estimated under free and submerged conditions using back vector machine models (SVM), spin multivariate adaptive regression (MARS), and multilayer artificial neural network (MLPNN) model. In free flow mode, the dimensionless parameters extracted from the dimensional analysis include the ratio of upstream flow to throat width and contraction ratio (throat width to channel width), and in the submerged state, in addition to these two parameters, the depth-to-throat width, and bottom-depth parameters upstream depth were used as input and the two-dimensional form of flow rate was used as the output of the models. The results showed that in free flow mode in the validation stage, the MARS model with statistical indices of R² = 0.985, RMSE = 0.008, MAPE = 0.87%, and the SVM model with statistical indices of R² = 0.971, RMSE = 0.0012, MAPE =1.376%, and MLPNN model with statistical indices of R² = 0.973, RMSE = 0.011, MAPE = 1.304% have modeled and predicted the flow rate. In the submerged state, the statistical indices of the developed MARS model were R² = 0.978, RMSE = 0.018, MAPE = 3.6%, and the statistical indices of the SVM model were R² = 0.988, RMSE = 0.014, 2%. MAPE = 4, and the statistical indicators of the MLPNN model were R² = 0.966, RMSE = 0.022, and MAPE = 5.7%. In the development of SVM and MLPNN models, radial kernel and hyperbolic tangent functions were used, respectively. [ABSTRACT FROM AUTHOR]

The current study aims to find the best methods of using remote sensing and supervised classification algorithms in long-term salinity monitoring of salinity changes in the Atabieh area with an area of 5000 hectares in the west of Khuzestan province. The procedure is based on the separation of different levels of saline soils utilizing information obtained from Landsat 7 and 8 satellite images (2001 to 2015) along with salinity data taken from the study area, and salinity indices including SI1, SI2, SI3, NDSI, IPVI, and VSSI. The results show the expansion of the saline zone trend in the soils of the study area, among which, soils with EC of more than 16 dS m-1 (very saline) have the highest frequency. The area of saline soils has increased significantly over the past 15 years, with a saline land area increasing by more than 90%. The percentage of salinity class is low (S1). According to this study, the only significant index in soil salinity at a 95% confidence level is the SI3 index, which has been able to have a good estimate of the increasing changes in soils in the region. The results of the supervised classification showed that the support vector machine (with an overall accuracy of 95.78 and a kappa coefficient of 0.89) is more accurate. After the vector machine method, the methods of minimum distance, maximum likelihood, and distance of Mahalanobis have the highest accuracy, respectively. Based on salinity maps obtained in years in 2001, 2005, 2010, and 2015, it can be said that the salinity rate in the whole of the study area was progressing and at the same time the salinity area in the middle and high classes increased decreased and on the other hand, the salinity area in the high class in 2001 gradually increased and distributed in 2015 throughout the region. [ABSTRACT FROM AUTHOR]

Predicting the status of power grid components, especially overhead lines in response to extreme weather event, plays an important role in planning pre-resource allocation, as well as restoring loads more quickly at a faster time which can improve the power grid resilience. In this paper, a Multi-dimensional Support Vector Machine algorithm is proposed to predict outage overhead lines assuming storm paths. In this model, dimensions are based on five distinct features: the distance from building, distance from tree, distance from center of storm, storm speed and cable type. The result of the proposed model is the classification of overhead lines status into two out and operational statuses implemented on the 33-bus IEEE network. The results show that the proposed model predicts outage lines with high accuracy. [ABSTRACT FROM AUTHOR]

The simulation of the rainfall-runoff process in the watershed has particular importance for a better understanding of hydrologic issues, water resources management, river engineering, flood control structures, and flood storage. In this study, to simulate the rainfall-runoff process, rainfall and discharge data were used in the period 1997-2017. After data qualitative control, rainfall, and discharge delays were determined using the coefficients of autocorrelation, partial autocorrelation, and cross-correlation in R Studio software. Then, the effective parameters and the optimum combination were determined by the Gamma test method and used to implement the model under three different scenarios in MATLAB software. Gamma test results showed that today's precipitation parameters, precipitation of the previous day, discharge of the previous day, and discharge of two days ago have the greatest effect on the outflow of the basin. Also, the Pt Qt-1 and Pt Pt-1 Qt-1 Qt-2 Qt-3 combinations were selected as the most suitable input combinations for modeling. The results of the modeling showed that in the support vector machine model, the Radial Base kernel Function (RBF) has a better performance than multiple and linear kernels. Also, the performance of the Artificial Neural Network model (ANN) is better than the Support Vector Machine model (SVM) with Radial Base kernel Function (RBF). [ABSTRACT FROM AUTHOR]

In this paper, a hybrid model for increasing the precision of the support vector machine classifier is proposed to detect stator windings short circuit fault detection in induction motors. The proposed method consists of three different phases, wherein the first phase the statistical features of a healthy and defective data set are extracted. The principal component analysis is used to reduce the dimensions of the obtained features. Then, different SVMs are constructed based on training data sets. To achieve a better result, the parameters of the SVM are determined by the fractionalorder chaotic particle swarm optimization algorithm. Finally, a hybrid model for combining SVMs with type-2 Fuzzy logic is implemented. The proposed approach is then applied on measured stator current data for stator winding short circuit fault detection in a three-phase induction motor with 2.2kW, 50Hz, 6 Pole. The average accuracy of 98.4% of the detection of stator winding error on laboratory data under different load conditions indicates the performance and validity of the proposed algorithm. [ABSTRACT FROM AUTHOR]

The prediction of local scouring as a dynamic and nonlinear phenomenon using methods of acceptable predictive capability has always been of interest to researchers. The shape of the bridge pier is one of the important factors in the formation and magnitude of the scour hole. In this paper, the scour depth of three bridge piers with cylindrical, sharp nose and rectangular shapes was predicted in two scenarios using the support vector machine algorithm with 395 field data obtained from the US Geological Survey and Froehlich (1988), based on different combinations of dimensionless parameters as the water attack angle (α), Froud number (Fr), the ration of pier length to width (l/b), and the ratio of mean sediment size to pier width (D50/b). The results of the study, while confirming the acceptable performance of the SVM algorithm for all piers in both scenarios, showed that in the first and second scenarios, the most optimal performance was related to the rectangular pier shape with correlation coefficient of 0.8702 and 0.8838, with and maximum Ds (DDR) values of 0.854 and 1.229 respectively, during the testing phase. The positive effect of increasing the number of data on the performance of the SVM algorithm was also confirmed by further probing the evaluation indicators. The results of the comparison pointed out the overestimation of the predicted scour depth values of absolute error between 11% to 35%. [ABSTRACT FROM AUTHOR]

Background and Aim Breast cancer is the abnormal cell growth in the breast. In both benign and malignant masses, there is rapid and high cell growth. Nowadays, due to the development of technologies, the diagnosis of diseases has become non-invasive and physicians attempts to diagnose the disease without surgery and based on internal organ images. Methods & Materials In this study, by using images prepared from the Digital Database for Screening Mammography (DDSM), a new method is proposed for detecting cancerous masses in the mammographic images using geometric features extraction and optimization of Support Vector Machine (SVM) parameters to classify breast cancer masses automatically. First, images were pre-processed and then boundaries were determined using threshold method. Next, morphological operators were used to improve these boundaries and the segmentation of images was carried out to classify cancerous masses. Finally, by using the SVM parameter optimization method, Grasshopper Optimization Algorithm (GOA), and 4-fold crossvalidation method, data were classified into two groups of benign and malignant (cancer) masses. Ethical Considerations Images from DDSM database were used in this research, all images are open access in this database. Results The accuracy, sensitivity and specificity values for applying the Radial Basis Function (RBF) kernel in SVM classifier (before optimization process) were obtained 97%, 100% and 96, respectively. After optimization of SVM parameters by the GOA, it was reported 100% for all accuracy, sensitivity and specificity indices for applying linear kernel function, indicating the high accuracy of the proposed method. The average values of accuracy, sensitivity and specificity indices for applying all three SVM kernel functions after optimization were 95.83, 100 and 94.81%, respectively. Conclusion The extracted geometrical features from breast cancer masses are highly efficient for model training and the diagnosis of breast cancer. The GOA could improve the overall accuracy of the proposed method by optimizing the SVM parameters. The results showed the higher performance of the proposed method compared to other methods. [ABSTRACT FROM AUTHOR]

Urban population growth and the continuous physical development of cities change the natural coverage of the earth and transform it into artificial cover and impervious surfaces. The dramatic increase of these surfaces yields negative consequences in many areas such as increasing surface runoff and flood risk, decreasing groundwater recharge, or intensifying the urban heat island effect. For these reasons, accurate estimation and monitoring of the trend of changes in these ranges is necessary. In this regard, remote sensing data are a cost-effective solution for the preparation and monitoring of impervious surfaces. The purpose of this study was to identify impervious urban surfaces using radar images. In the present study, the textural properties of the Gray Level Co-occurrence Matrix (GLCM) were evaluated using maximum likelihood classification methods, artificial neural network, and support vector machine on Sentinel-1 radar image to determine impervious surfaces of Bandar Abbas city. The overall accuracy of 97.00%, 98.14%, 98.40% and Kappa coefficient of 0.95, 0.97, and 0.97, respectively, for maximum likelihood classification, artificial neural network, and support vector machine, indicated the appropriateness of the utilized methods for detecting impervious urban surfaces. For extracting urban surface information, spectral feature classification algorithms are mostly used. This causes a large amount of useful spatial information such as texture to be ignored in the classification images. Given that SAR images are sensitive to the geometrical properties of urban surfaces, impervious urban surfaces can be accurately detected by using textural properties of radar imagery, which has not been addressed so far. [ABSTRACT FROM AUTHOR]

Alzheimer's is a type of brain dementia that gradually reduces mental abilities of the patient. The lack of memory, decision-making disorder, and mistakes in choosing the correct vocabulary are the early symptoms of Alzheimer's disease. Therefore, extensive studies have been conducted on the diagnosis of Alzheimer's disease using the noninvasive speech signal recognition method. Identifying of Alzheimer's disease is dependent on culture and language, speech content, gender, age, accent, and many other factors. Therefore, Alzheimer's speech signal has been studied in various languages. The purpose of this paper is to recognize Alzheimer's patients from healthy people by the use of their speech signal processing in Persian using the combination of time, frequency, and frequency-temporal features. In this paper, after pre-processing, the speech features extracted using the wavelet packet as a frequency-temporal feature next to Mel frequency Cepstral coefficients, zero crossing rate, spectral roll off, band width, root mean square and spectral centroid frequency. Finally, the extracted features have been classified by the support vector machine which achieves recognition precision of 96% on Persian healthy and Alzheimer's speaker experiments. The acceptable results demonstrate the applicability of the proposed non-invasive and low-cost algorithm for the diagnosis of Persianspeaking Alzheimer's patients. [ABSTRACT FROM AUTHOR]

The growing use of internet and the existence of vulnerable points in networks have made the use of intrusion detection systems as one of the most important security elements. Intrusion detection is essentially a classification problem and it is the identification of effective features such as important issues in the classification This paper presents a novel method for selecting effective features in network intrusion detection based on an estimation of distribution algorithm that uses a probabilistic dependency tree to identify important interactions between features. To evaluate the performance of the proposed method, the NSL- KDD dataset is used, in which the packets are divided into five normal types and intrusive types of DOS, U2R, R2L and Prob. The performance of the proposed algorithm has been compared alone and in combination with other feature selection algorithms such as forward selection, backward selection and genetic algorithm. Moreover, the effect of algorithm parameters like population size on intrusion detection accuracy is tested. Based on this analysis and also considering the intra-class accuracy of different feature selection methods studied in this paper, an effective subset of features for intrusion detection is identified. [ABSTRACT FROM AUTHOR]

Land use/cover maps are the basic inputs for most of the environmental simulation models; hence, the accuracy of the maps derived from the classification of the satellite images reduces the uncertainty in modeling. The aim of this study was to assess the accuracy of the maps produced by machine learning based on classification methods (Random Forest and Support Vector Machine) and to compare them with a common classification method (Maximum Likelihood). For this purpose, the image of the OLI sensor of Landsat 8 for the study area (Sattarkhan Dam’s basin in the Eastern Azerbaijan) was used after the initial corrections. Five land uses including urban, irrigated and rain-fed agriculture, range and water body were considered. For conducting the supervised classification, ground truth data were used in two sets of educational (70% of the total) and test (30%) data. Accuracy indexes were used and the McNemar test was employed to show the significant statistical difference between the performances of the methods. The results indicates that the overall accuracy of Support Vector Machine, Random Forest, and Maximum Likelihood methods was 96.6, 90.8, and 90.8 %, respectively; also the Kappa coefficient for these methods was 0.93, 0.81 and 0.83, respectively. The existence of a significant statistical difference at the 95% confidence between the performances of the Support Vector Machine algorithm and the other two algorithms was confirmed by the McNemar test. [ABSTRACT FROM AUTHOR]

A labyrinth weir is a nonlinear weir folded in the plan-view which increases the crest length and the flow rate for a given channel width and an upstream flow depth. Nowadays, a labyrinth weir is an attractive alternative for those weirs that have a problem in passing the probable maximum flood. The three-dimensional flow pattern and unlimited geometric parameters provide a major challenge to the designers of these weirs. The present study aimed at determining discharge coefficients of sharp-crested triangular labyrinth weirs using the support vector machine (SVM). The results were compared with the experimental data. For this purpose, 123 laboratory test data including geometric and hydraulic parameters such as vertex angle (θ), magnification ratio (L/B), head water ratio (h/w), Froude number (Fr), Weber Number (We) and Reynolds number (Re) were used. The results showed that the SVM-based model produced the most accurate results when only three geometric parameters, e.g. (h/w, θ, L/B), were introduced as the input parameters (R2 = 0.974, Root mean square error [RMSE] = 0.0118, mean absolute error [MAE] =0.0112 and mean normal error [MNE] =0.017 for the test stage). Also, for these weirs, polynomials linear and nonlinear regression equations were presented. Finally, the discharge coefficient of sharp-crested triangular labyrinth weirs based on the Rehbock equation was evaluated and compared with the SVM using nonlinear and linear regression methods. [ABSTRACT FROM AUTHOR]

This paper presents a new approach for islanding detection based on fast STransform, in distribution networks with distributed generators. In this method, at first, S-Transform of voltage and current at point of common coupling are calculated; then, features of normal and islanding operation are extracted using S matrix and frequency contours. The features are extracted using several simulations; considering load switching, motor load switching, transient faults, and islanding conditions with different power mismatches for local load. Finally, support vector machine has been proposed for classifying the extracted features to detect the islanding. For more studies, a distribution system is simulated using PSCAD/EMTDC and feature vectors corresponding to different situations of islanding and normal operation are extracted and used for training and test of support vector machine classifier. To evaluate the performance of proposed scheme, the obtained results are compared with results of other methods. Comparison of the results shows that the proposed method has higher accuracy and speed in islanding detection. Also, the proposed approach is effective, fast and would not be affected by noise during classifying of conditions. [ABSTRACT FROM AUTHOR]