Search

Accurate and up-to-date information on urban built-up areas is significant for managing urban growth and development. Earth Observation (EO) data are valuable sources for meeting this demand. However, the extraction of urban built-up areas from EO data is challenging due to the limitations of EO data sources. To overcome this challenge, this study follows an approach that assesses the performance of optical (Sentinel-2), radar (Sentinel-1) and fused (Sentinel-1 and Sentinel-2) data to extract urban built-up areas using machine learning algorithms including Random Forest (RF), K-Nearest Neighbors (KNN) and KDTree KNN. The results were statistically analyzed by considering the Overall Accuracy (OA) and kappa coefficient. In addition, 15 cm GSD (Ground Sample Distance) aerial photography of the study area was used to validate the results. According to the results, Sentinel-2 produced better representation and accuracy of urban built-up areas than Sentinel-1 and even the fused image. Regarding to machine learning algorithms classification performance, RF performed better in both OA and Kappa coefficient along all datasets. The research findings can have significant implications for various domains, such as urban planning, land use management and open avenues for further comparisons of different EO data sources and machine learning algorithms for built-up areas extraction. [ABSTRACT FROM AUTHOR]

Cyber-attacks using malicious uniform resource locator (URL) propagation are very common and serious. Statistics indicate that there is a need to research and apply techniques and methods for identifying and preventing malicious URLs. The main objective of this research is to train machine learning models on selected dataset to predict phishing websites based on URL-related features. The accuracy level of each model is measured and compared. Finally, the best performing model will be used to develop a web application that provide internet users with an easy way to check suspicious URLs. We have used five different machine learning models to classify URLs as legitimate or phishing, these models are eXtreme Gradient Boosting (XGBoost), k-nearest neighbors (KNN), support vector machine (SVM), Decision Tree, and Random Forest. Finally, we used Voting Classifier to combine the work of Random Forest (RF) algorithm with other two models, Gaussian Naive Bayes, and Logistic Regression, to check if we can increase the accuracy of RF as suggested in the literature, but we found that the accuracy of RF alone was higher than the accuracy of the combined models. This project can be implemented as a browser extension or mobile application to classify suspicious URLs to legitimate or phishing with the use of the saved model. [ABSTRACT FROM AUTHOR]

Current measurement systems based on the IEEE-1159 standard have some limitations and robustness problems under noisy and fast-changing conditions. Besides, applying different methods for each Power Quality Disturbance (PQD) to every window is required but time-consuming and not feasible. Therefore, different kinds of two-stage methods, Detection and Classification (D&C), have been improved in many studies. Then, the required measurement can be performed to define disturbance. For this purpose, a new approach based on features of subcomponents with Machine Learning Algorithms (MLAs) to detect and classify PQDs is proposed. 21-class dataset including single and multiple PQDs under different noisy conditions was prepared randomly. Of this dataset, determined features were extracted and some of these were selected. Then, selected features were trained and tested with some MLAs in a workstation. Results obtained from comparative MLAs and the other classification methods show that the best MLA with related features is Random Forest with 96.97% while LightGBM, k-Nearest Neighbors, and XGBoost 96.85%, 96.73%, and 92.82% accuracy, respectively. Because the selected features, optimized parameters, and the related MLA were obtained by investigating for features provided from the PQDs in the whole parameter space, this approach brings the advantages of high accuracy, low D&C complexity, and computing load. [ABSTRACT FROM AUTHOR]

In this article, we consider the k-Nearest Neighbors (k-NN) method in a single index regression model when the explanatory variable is valued in functional space in the setting of the quasi-association dependence condition. The main result of this work is the establishment of the asymptotic distribution for the k-NN kernel single index estimator. These results are established under fairly general conditions on the underlying models. As an application, the asymptotic confidence bands for the regression model based on the single-index model are presented. Some simulation studies are carried out to show the finite sample performances of the k-NN estimator. [ABSTRACT FROM AUTHOR]

Machine learning techniques have been widely used in the oil and gas industry to improve the qualitative and quantitative characterization of subsurface reservoirs. Because rock properties are strongly influenced by lithological and sedimentological information, lithofacies classification is an important step in 3D reservoir modeling. The aim of this study is to use supervised classification algorithms to predict the spatial distribution pattern of lithofacies classes using borehole and seismic data. In this study, lithofacies classes are distributed away from the wells using a machine-learning classifier. Seismic data attributes extracted from well locations are utilized as training data features in various supervised classification algorithms. Machine learning classifiers trained and evaluated for lithofacies classification include K-nearest neighbors, support vector machine, Gaussian naive Bayes, decision tree, Gradient Boosting, and Random Forests. A number of parameters are optimally determined in order to achieve the highest value of classification accuracy in the model. Comparing machine learning classifiers based on evaluation metrics reveals that ensemble-based decision tree approaches such as Random Forests and Gradient Boosting are the most effective for supervised classification. The results are validated using testing data and have an 80% classification accuracy. The predicted volume of lithofacies classes contributes to improved 3D reservoir modeling for the pre-salt carbonate reservoir. [ABSTRACT FROM AUTHOR]

Chrysanthemum tea is popular because of its health benefits, although different varieties vary in effectiveness and role. The rapid identification of the variety is critical for consumers and the tea industry. This study employed near-infrared (NIR) spectroscopy in combination with feature extraction methods to characterize chrysanthemum tea varieties. To improve the recognition accuracy, a fuzzy improved pseudoinverse linear discriminant analysis (FIPLDA) algorithm was employed to extract discriminant information from the NIR spectra of chrysanthemum tea that were preprocessed using the Savitzky-Golay (SG) algorithm. Next, the dimensionality of the data was reduced by principal component analysis (PCA). The spectral features were extracted by pseudoinverse linear discriminant analysis (PLDA), improved pseudoinverse linear discriminant analysis (IPLDA), and FIPLDA. Five chrysanthemum tea samples were classified by k-nearest neighbor (KNN) and support vector machines (SVM), respectively. The results show that KNN performed better than SVM in classification, and the classification accuracy of FIPLDA-KNN reached 98.33%. Hence, the combination of NIR spectroscopy with FIPLDA suitable for distinguishing chrysanthemum tea. [ABSTRACT FROM AUTHOR]

The weighted nearest neighbors (WNN) estimator has been popularly used as a flexible and easy-to-implement nonparametric tool for mean regression estimation. The bagging technique is an elegant way to form WNN estimators with weights automatically generated to the nearest neighbors (Steele 2009; Biau, Cérou, and Guyader 2010); we name the resulting estimator as the distributional nearest neighbors (DNN) for easy reference. Yet, there is a lack of distributional results for such estimator, limiting its application to statistical inference. Moreover, when the mean regression function has higher-order smoothness, DNN does not achieve the optimal nonparametric convergence rate, mainly because of the bias issue. In this work, we provide an in-depth technical analysis of the DNN, based on which we suggest a bias reduction approach for the DNN estimator by linearly combining two DNN estimators with different subsampling scales, resulting in the novel two-scale DNN (TDNN) estimator. The two-scale DNN estimator has an equivalent representation of WNN with weights admitting explicit forms and some being negative. We prove that, thanks to the use of negative weights, the two-scale DNN estimator enjoys the optimal nonparametric rate of convergence in estimating the regression function under the fourth-order smoothness condition. We further go beyond estimation and establish that the DNN and two-scale DNN are both asymptotically normal as the subsampling scales and sample size diverge to infinity. For the practical implementation, we also provide variance estimators and a distribution estimator using the jackknife and bootstrap techniques for the two-scale DNN. These estimators can be exploited for constructing valid confidence intervals for nonparametric inference of the regression function. The theoretical results and appealing finite-sample performance of the suggested two-scale DNN method are illustrated with several simulation examples and a real data application. [ABSTRACT FROM AUTHOR]

Since dental materials are worn down over time and eventually need to be replaced. Resin composites are frequently employed as dental restorative materials. By employing the in-vitro test findings of the pin-on-disc tribometer [ASTM G99-04], the goal of this study is to evaluate the capability of three different machine learning (ML) models in analyzing the wear of dental composite materials when immersed in chewable tobacco solution. Four distinct dental composite material samples are used in this investigation, and after being dipped in a chewing tobacco solution for a few days, the samples are taken out and subjected to a wear test. Three different ML models (MLP, KNN, XGBoost) have been chosen for predicting the wear of dental composite specimens. XGBoost ML model yields an R2 value of 0.9996 and it performs noticeably better than the other approaches. [ABSTRACT FROM AUTHOR]

Owing to the predominant role of Machine Learning(ML) across domains, it is being introduced at multiple levels of education, including K-12. Researchers have leveraged games, augmented reality and other ways to make learning ML concepts interesting. However, most of the existing games to teach ML concepts either focus on use-cases and applications of ML instead of core concepts or directly introduce ML terminologies, which might be overwhelming to school students. Hence, in this paper, we propose ML-Quest, a game to incrementally present a conceptual overview of three ML concepts: Supervised Learning, Gradient Descent and K-Nearest Neighbor (KNN) Classification. The game has been evaluated through a controlled experiment, for its usefulness and player experience using the TAM model, with 41 higher-secondary school students. Results show that students in the experimental group perform better in the test than students in the control group, with 5% of students in the experimental group scoring full marks. However, none of the students in the control group could score full marks. The survey results indicate that around 77% of the participants who played the game either agree or strongly agree that ML-Quest has made their learning interactive and is helpful in introducing them to ML concepts. [ABSTRACT FROM AUTHOR]

Wastewater pollution caused by organic dyes is a growing concern due to its negative impact on human health and aquatic life. To tackle this issue, the use of advanced wastewater treatment with nano photocatalysts has emerged as a promising solution. However, experimental procedures for identifying the optimal conditions for dye degradation could be time-consuming and expensive. To overcome this, machine learning methods have been employed to predict the degradation of organic dyes in a more efficient manner by recognizing patterns in the process and addressing its feasibility. The objective of this study is to develop a machine learning model to predict the degradation of organic dyes and identify the main variables affecting the photocatalytic degradation capacity and removal of organic dyes from wastewater. Nine machine learning algorithms were tested including multiple linear regression, polynomial regression, decision trees, random forest, adaptive boosting, extreme gradient boosting, k-nearest neighbors, support vector machine, and artificial neural network. The study found that the XGBoosting algorithm outperformed the other models, making it ideal for predicting the photocatalytic degradation capacity of BiVO4. The results suggest that XGBoost is a suitable model for predicting the photocatalytic degradation of wastewater using BiVO4 with different dopants. [ABSTRACT FROM AUTHOR]

Identifying faults in the photovoltaic (PV) arrays is very much essential in improving the PV system's safety and reliability. Solar PV arrays operate with non-linear characteristics, installed with maximum power point trackers (MPPT's), and blocking diodes cause mismatch levels. Line-to-line and line-to-ground faults are identified, and the faulted circuits are isolated by means of over current protection devices (OCPD) and ground fault protection devices (GFPD). In order to improve the accuracy of fault detection, artificial intelligence (AI)-based techniques like Fuzzy inference, wavelet, support vector machine, and k-nearest neighbors are used. The drawback of AI-based techniques are (1) requirement of large dataset for effective fault identification and also show incompatibility if there is low irradiation and (2) require a larger number of voltage and current sensors. An experimental setup of 160 W, 4 × 4 solar PV array having PV modules (SPB) is subjected to different fault conditions (CS), and the faults are identified using the minimum number of sensors. The faults that are not detected by the conventional methods are detected using this proposed method, and the power gain due to the fault identification is around 152% which is 97 W in the PV array. [ABSTRACT FROM AUTHOR]

To address the problem of low quality of the outlier detection results caused by the irregular spatial distribution of crowdsourced bathymetric data, an intelligently optimized 3D-DBSCAN method is hereby proposed for detecting outliers in crowdsourced bathymetric data. Firstly, the potential of the classic DBSCAN algorithm in clustering irregularly distributed data has been explored for three-dimensional space and upgraded to the 3D-DBSCAN algorithm. Meanwhile, three key parameters affecting the 3D-DBSCAN algorithm results have been determined, namely, the minimum number of objects in the neighborhood, the horizontal neighborhood radius and the vertical neighborhood radius. Then, the abilities of the K-nearest neighbor method and the genetic algorithm in setting the initial range of parameters and searching for optimal values have been comprehensively utilized. The optimal combination solution vector of the key parameters of 3D-DBSCAN under different distribution conditions has been adaptively acquired by defining and calculating the silhouette coefficient index. Finally, spatial clustering analysis on crowdsourced bathymetric data and intelligently detect outliers has been conducted by using the optimal combination solution vector as the input of the 3D-DBSCAN model parameters. The experimental results show that the proposed method can not only detect outliers in crowdsourced bathymetric data after adaptively optimizing the parameters, but also detect outliers by analyzing the distance relationship between points in the 3D space, thereby overcoming the limitations of traditional methods in detecting water depth anomalies at specific positions and irregularly distributed anomalies. [ABSTRACT FROM AUTHOR]

This research integrates machine learning (ML) and Six Sigma's Define, Measure, Analyze, Improve, and Control (DMAIC) methodology to address these issues. The study details the selection and utilization of ML techniques, including Linear Regression (LR), Artificial Neural Network (ANN), Decision Tree (DT), K-nearest neighbors (KNN), and Cluster Analysis (CA). Implemented at the Innovative Plastic Manufacturing Company in Egypt, this research enhances the consistency of plastic bottle production by addressing issues such as surface marks, flashes, bubbles, and variations in liter capacity. Integrating Six Sigma with ML techniques reduces the average defect rate from approximately 67.8%. It elevates the Sigma level from 3.14 to 4.30, reducing material over-consumption costs from 5% to 1.7% of total manufacturing expenses. Notably, the KNN model achieves the best results for defect testing, with an R-squared value of 98.8%. These methodologies lead to cost reduction, increased competitiveness, and improved product quality when implemented. [ABSTRACT FROM AUTHOR]

With the continuous development of education, personalized learning has attracted great attention. How to evaluate students' learning effects has become increasingly important. In information technology courses, the traditional academic evaluation focuses on the student's learning outcomes, such as "scores" or "right/wrong," which seldom reflects the development of students' cognitive level and lacks effective diagnostic information. This article proposes a non-parametric multi-level scoring cognitive diagnosis method based on the KNN and the characteristics of information technology courses named the EW-KNN (E-weight K-Nearest Neighbor). Compared with the KNN, the EW-KNN improved two key points. One is that it takes the number of IRP (Ideal Response Pattern) as the K value to adapt to different types of tests. The other is that the nearest neighbor distance is introduced to solve the problem of misjudgment of the categories. The Monte Carlo simulation method is used to test its performance. The results indicate that the EW-KNN has a higher accuracy rate and is suitable for information technology courses. Furthermore, the method is applied in information technology course to make a cognitive diagnosis of 120 students of high school in Shanghai. Results demonstrate that the EW-KNN can accurately diagnose each student's cognition levels and knowledge structure accurately. [ABSTRACT FROM AUTHOR]

The cognitive workload is a key to developing a logical and conscious thinking system. Maintaining an optimum workload improves the performance of an individual. The individuals' psycho-social factors are responsible for creating significant variability in the performance of a task, which poses a significant challenge in developing a consistent model for the classification of cross-task cognitive workload using physiological signal, Electroencephalogram (EEG). The primary focus of the proposed work is to develop a robust classification model CARNN, by employing the concatenated deep structure of distributed branches of convolutional neural networks with residual blocks through identity mappings, and recurrent neural network with an attention mechanism. EEG data is divided into milliseconds duration overlap segments. The segmented EEG data is converted into images using Gabor decomposition with two spatial frequency scales and four orientations and supplied as input to CARNN. The images are formed by interlacing the respective left and right electrode data to capture the data variations effectively. Efficient feature aggregation with learning of spatial and temporal domain discriminative features through Gabor decomposed data images improve the training of CARNN. CARNN achieves outstanding performance over traditional classifiers; support vector machine, k-nearest neighbor (KNN), ensemble subspace KNN and the pre-trained networks; AlexNet, ResNet18/50, VGG16/19, and Inception-v3. The proposed method results in 94.2%, 92.5%, 95.9%, 92.8%, 94.3% classification accuracy, specificity, sensitivity, precision, and F1-score, respectively. Two visual task levels apart in their complexity are used for cross-task classification of cognitive workload. The proposed method is validated on raw EEG data of 44 participants. [ABSTRACT FROM AUTHOR]

Ultrasound scanning has been used as the preliminary diagnosis tool all over the world. The ultrasound data are being analyzed by tele-radiologists. It lacks online availability. The drawbacks of tele-radiology have been overcome by using computer-aided diagnosis. As an aid to this, the Random Forest Classifier has been used here for detecting kidney abnormalities. The initial pre-processing stage filters the speckle noise existing in the input kidney image. Then feature extraction has been performed. Image categorization as normal, cyst and stone has been done with Random Forest Classifier. Then the performance is evaluated by comparing it with K-nearest neighbor classifier and support vector machine. From experimentation, it is observed that the accuracy and F-measure values of Random Forest Classifier range high when compared with other classifiers due to the continuous split of the trees until accurate categorization is done. Simulation and Implementation have been done using Modelsim 6.4a and Xilinx Spartan-6 FPGA board. [ABSTRACT FROM AUTHOR]

Self-supervised metric learning has been a successful approach for learning a distance from an unlabeled dataset. The resulting distance is broadly useful for improving various distance-based downstream tasks, even when no information from downstream tasks is used in the metric learning stage. To gain insights into this approach, we develop a statistical framework to theoretically study how self-supervised metric learning can benefit downstream tasks in the context of multi-view data. Under this framework, we show that the target distance of metric learning satisfies several desired properties for the downstream tasks. On the other hand, our investigation suggests the target distance can be further improved by moderating each direction's weights. In addition, our analysis precisely characterizes the improvement by self-supervised metric learning on four commonly used downstream tasks: sample identification, two-sample testing, k-means clustering, and k-nearest neighbor classification. When the distance is estimated from an unlabeled dataset, we establish the upper bound on distance estimation's accuracy and the number of samples sufficient for downstream task improvement. Finally, numerical experiments are presented to support the theoretical results in the article. for this article are available online. [ABSTRACT FROM AUTHOR]

A new machine learning-based method to classify the different transient events in distributed generation (DG) system has been proposed in this article. An existing hybrid DG-based network which consists of three microgrids (MGs), i.e. thermal, wind, and solar power, is used as test network to create transient conditions for both the islanding and grid-connected circumstances. The transient case studies include the symmetrical and unsymmetrical fault at distribution line, intentional islanding, variation of power demand, switching of capacitor bank, addition of nonlinear load, motor starting condition, etc. This recommended methodology starts with generating the sampled voltage signals of three different phases of different locations, and each signal has been decomposed using discrete wavelet transform. The significant features are extracted from the computed energy values of detailed wavelet coefficient for co-training of fine K-nearest neighbor (KNN) and ensemble KNN classification in the following stage. The results and the performance indices of the trained classifiers prove that the proposed method has been detected and classified all the transient events with 98% accuracy. Such type of multiple transient event classification in MG by a single algorithm is truly beneficial with respect to the power quality issues of modern power system. [ABSTRACT FROM AUTHOR]

In this article, a nonparametric lack-of-fit test of nonlinear regression in presence of heteroscedastic variances is proposed. We consider regression models with a discrete or continuous response variable without distributional assumptions so that the test is widely applicable. The test statistic is developed using a k-nearest neighbor augmentation defined through the ranks of the predictor variable. The asymptotic distribution of the test statistic is derived under the null and local alternatives for the case of using fixed number of nearest neighbors. The parametric standardizing rate is achieved for the asymptotic distribution of the proposed test statistic. This allows the proposed test to have faster convergence rate than most of nonparametric methods. Numerical studies show that the proposed test has good power to detect both low and high frequency alternatives even for moderate sample size. The proposed test is applied to an engineering data example. [ABSTRACT FROM AUTHOR]

A method for discriminating Salvia miltiorrhiza from different geographical origins was developed using laser-induced breakdown spectroscopy (LIBS) with convolutional neural network (CNN). The LIBS spectra of Salvia miltiorrhiza samples from six geographical locations were preprocessed with the maximum minimum normalization method. The classification model for discriminating these samples was developed by using a one-dimensional convolutional neural network. The discrimination accuracy of the developed CNN model reached 97.09%. Compared with support vector machine and k-nearest neighbor methods, the CNN model showed higher discrimination accuracy. The results demonstrate that the combination of LIBS and CNN is suitable for discriminating Salvia miltiorrhiza from different geographical locations. [ABSTRACT FROM AUTHOR]

Performance assessment of Small and Medium-Sized Water Systems (SMWSs) is important for operational, tactical, and strategic decision-making. In this study, a performance assessment method has been developed and applied to five drinking water, three wastewater, and two stormwater utilities using 39, 30 and 27 Key Performance Indicators (KPIs) in a semi-arid region. The KPIs were aggregated to determine a performance index using a Technique for Order of Preference by Similarity to Ideal Solution method. K-nearest neighbors and penalty methods were used to estimate missing KPIs data. The results indicated that only two drinking water utilities and one wastewater utility had been rated as 'high' performance. None of the utilities in stormwater performance was rated as 'high'. The developed method can assist decision-makers in evaluating SMWSs performance holistically, build operational management strategies, and identify necessary interventions in overcoming water systems challenges across each urban water system component. [ABSTRACT FROM AUTHOR]

A Deep Long Short-Term Memory (DeepLSTM) Deep Learning model for the classification of personality traits using Electroencephalogram (EEG) signals is developed in this study. The objective is to assess the efficiency of the DeepLSTM model in classification. The publicly available ASCERTAIN EEG dataset is used, which uses the Big Five Factor model for predicting personality. We also evaluated DeepLSTM model performance to that of existing state-of-the-art based machine learning classifiers such as Multilayer Perceptron (MLP), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and LibSVM. The proposed model outperforms the existing classifiers for the 70-30 partitioning approach, with a maximum classification accuracy of 90.32%. [ABSTRACT FROM AUTHOR]

Convex clustering has recently garnered increasing interest due to its attractive theoretical and computational properties, but its merits become limited in the face of high-dimensional data. In such settings, pairwise affinity terms that rely on k-nearest neighbors become poorly specified and Euclidean measures of fit provide weaker discriminating power. To surmount these issues, we propose to modify the convex clustering objective so that feature weights are optimized jointly with the centroids. The resulting problem becomes biconvex, and as such remains well-behaved statistically and algorithmically. In particular, we derive a fast algorithm with closed form updates and convergence guarantees, and establish finite-sample bounds on its prediction error. Under interpretable regularity conditions, the error bound analysis implies consistency of the proposed estimator. Biconvex clustering performs feature selection throughout the clustering task: as the learned weights change the effective feature representation, pairwise affinities can be updated adaptively across iterations rather than precomputed within a dubious feature space. We validate the contributions on real and simulated data, showing that our method effectively addresses the challenges of dimensionality while reducing dependence on carefully tuned heuristics typical of existing approaches. for this article are available online. [ABSTRACT FROM AUTHOR]

Parkinson's disease (PD) is one of the most widespread neurological disorders associated with nerve damage without definitive treatment. Impairments, such as trembling and slowing down in hand movements are among the first symptoms. For this purpose, in this study, machine learning (ML)-based models were developed by using keyboard keystroke dynamics. According to patients' drug use status, disease severity, and gender, we created 14 different sub-datasets and extracted 378 features using raw keystroke data. We developed alternative models with Support Vector Machines (SVM), k-Nearest Neighbors (kNN), and Random Forest (RF) algorithms. We further used Minimum Redundancy Maximum Relevance (mRmR), RELIEF, sequential forward selection (SFS), and RF feature selection methods to investigate prominent features in distinguishing PD. We developed ML models that jointly use the most popular features of selection algorithms (feature ensemble [FE]) and an ensemble classifier by combining multiple ML algorithms utilizing majority vote (model ensemble [ME]). With 14 different sets, FE and ME models provided accuracy (Acc.) in the range of 91.73 − 100% and 81.08 − 100%, respectively. [ABSTRACT FROM AUTHOR]

In this article, we propose a three-dimensional (3D) face recognition approach for depth data captured by Kinect based on a combination of speeded up robust features (SURF) and k-nearest neighbor (KNN) algorithms. First, the shape index maps of the preprocessed 3D faces of the training gallery are computed, then the SURF feature vectors are extracted and used to form the dictionary. In the recognition process, we propose an improved KNN classifier to find the best match. The evaluation was performed using CurtinFaces and KinectFaceDB data sets, achieving rank-1 recognition rates of 96.78% and 94.23%, respectively, when using two samples per person for training. [ABSTRACT FROM AUTHOR]

Because of its simplicity and high efficiency, logistic regression algorithm has been widely applied in the field of data mining. However, logistic regression classifiers always show poor performance when the categories are imbalanced. In this paper, a new algorithm named K ′ KLR ( k ′ times k-means logistic regression) is proposed to deal with the imbalanced classification problem. Compared to other traditional methods, the effectiveness of the proposed algorithm is verified through experiments based on simulated data and real data. [ABSTRACT FROM AUTHOR]

This paper proposes a machine learning method for automatically detecting and identifying facies from digital images of the oil well core. The method is based on artificial neural networks, specifically, pre-trained deep convolutional neural networks, improved using histogram of oriented gradients and local binary pattern methods. AlexNet, Inception, MobileNet, Xception, and DenseNet extract features from images. An algorithm based on the behavior of fireflies selects the most informative features. The k-nearest neighbors, support vector machines, and random forest methods are considered as classifiers. The proposed approach uses a model that has been trained and tested on a core rock samples dataset of > 23,000 images related to four facies as coal, sandstone, siltstone, and shale. The experimental results show that the proposed HOG + LBP + DenseNet model combined with a random forest is better than support vector machines and k-nearest neighbors methods regarding facies recognition from core photographs and the precision (95.5%), recall (96.46%), and F-measure (95.98%) metrics. [ABSTRACT FROM AUTHOR]

The aroma indicator is an important factor in evaluating the quality and grade of black tea and is vital in preventing fraud and reducing financial losses. A rapid method for the characterization of Congou black tea is described using an electronic nose (EN) with grey wolf optimization (GWO) and chemometrics. First, the aroma qualities of the tea were determined by the EN for 700 samples of seven grades and converted into 10 characteristic values. The GWO algorithm was used to optimize the aroma indicators of the EN sensors. Discrimination models based upon the optimized odor sensor features and partial least squares discriminant analysis, k-nearest neighbor (KNN), extreme learning machine, and support vector machine were created to assess the quality of the samples. The results showed that the KNN model constructed with the five odor sensor response values optimized by the GWO algorithm performed best with an accuracy of 97.85%. The results demonstrate that the EN is suitable for the rapid assessment of the quality and grades of Congou black tea products. [ABSTRACT FROM AUTHOR]

Early screening of cancer plays a vital role in successful treatment. Normally, the temperature of breast surface changes due to the size and location of the underlying tumor. Temperature patterns can be used to estimate the size and location of tumor by an inverse approach using machine learning algorithms. The present study aims to provide an efficient machine learning model that can be relied on to predict the size and location of the tumor using numerical thermal images. There is no availability of actual thermal images labeled with the size and location of the tumor. Consequently, successive numerical simulations are used to develop the numerical thermal image dataset with a three-dimensional breast model solved with COMSOL Multiphysics. The obtained numerical thermal images are uniquely labeled with the corresponding size and location of the tumor and trained using various machine learning regression models such as linear, support vector, K-nearest neighbor, and decision tree regression models. The results are analyzed using parity plots and mean absolute error. The present study found that the decision tree regression model outperformed the other machine learning models and improved the estimation accuracy by rejecting the numerical thermal images which are having slight variation in surface temperature. [ABSTRACT FROM AUTHOR]

In this communication, a metasurface-loaded ceramic-based dual port antenna is designed and raised via artificial neural networks (ANN), k-nearest neighbor (KNN) and XG Boost. With the assistance of these algorithms, a reliable and flexible framework is obtained to predict the optimized design parameters of the proposed radiator. To confirm the accuracy of these ML algorithms, predicted outcomes is compared with outcomes obtained from HFSS and fabricated prototype. The predicted results are very well matched with practical conclusion. The designed radiator operates from 2.55 to 3.2 GHz, along with supporting circularly polarized waves from 2.67 to 3.1 GHz. These appearances make the proposed aerial suitable for the sub-6 GHz frequency band. [ABSTRACT FROM AUTHOR]

As macroscopic Breast Cancer (BC) imaging diagnostics do not provide satisfied data, microscopic image analysis is used to get optimum cell-nuclei segmentation for image categorization. The paper presents a Novel Hybrid Segmentation Method which uses the energy detail coefficients attained from Non-Subsampled Contourlet Transform for cell-nuclei segmentation of BC histopathological images. NHSM solves the local inhomogeneities problem while preserving all necessary details. The system evaluation is done using training dataset of 98 BC histopathological images and testing dataset of 26 images. Total of 25 segmented features was extracted for classification using Decision Trees, k-Nearest Neighbor, multilayer perceptron, Multiclass-Support Vector Machine (M-SVM). The proposed NHSM segmentation results were undergone through both subjective and objective performance evaluation. Result evaluation proved that NHSM along with Multi Class-Support Vector Machine classifier provides highly precise and accurate classified result for BC tissue images compared to state-of-the-art segmentation methods. Also, the edge detection model initialization is highly robust and insensitive. [ABSTRACT FROM AUTHOR]

Machine learning technique involves a significant amount of time and the model performance reduces in multi-dimensional datasets because of redundant features. Feature selection is a significant step in machine learning and involves the selection of a subset of relevant data feature from larger data feature, which further enhances model performance by simplification. Crop yield prediction is a significant application of machine learning, and feature selection plays a crucial role in it. To predict paddy yield accurately, weather, soil, and crop attributes are essential factors. Therefore, feature selection techniques are employed to identify relevant and non-redundant attributes from a larger dataset, which simplifies the prediction model. In this study, an ensemble feature selection method is proposed that selects an optimized subset of attributes by combining various attribute subsets based on mutual information between attributes and between attributes and the class. The proposed ensemble approach is validated using five classification techniques, including K-nearest neighbor, Random Forest, Support Vector Machine, Naive Bayes, and Bagging. Several evaluation metrics such as Accuracy, Error rate, Kappa, Precision, Recall, Specificity, and F1 score are used to assess the performance of the ensemble approach and compare it with other feature selection techniques. The experimental results indicate that the proposed ensemble approach with the Random Forest classifier outperforms other classifiers, with Accuracy and Error rate values of 0.9491 and 0.0509, respectively. [ABSTRACT FROM AUTHOR]

We consider the problem of nonparametric estimation of certain conditional models when the explanatory variable is functional. The models studied are the Conditional Expectation (CE), the Conditional Distribution Function (CDF) and the Conditional Probability Density (CPD). The estimators are constructed by combining the local linear method and the k-Nearest Neighbors (kNN) smoothing approach. For each of the three models, we define an optimal estimator associated with the best number of neighbors chosen using two bandwidth selection procedures which are the cross-validation criterion and the bootstrap smoothing rule. The asymptotic properties of these optimal kNN Local Linear Estimators ( k NN- LLE) are established under some standard conditions. The performances of the finite samples of these estimators are then examined and compared through a Monte-carlo study. In addition, an application on the riboflavin content in the yogurt using the Near-infrared curves is carried out to demonstrate the usefulness of the models proposed in the point-wise prediction as well as for the predictive interval. [ABSTRACT FROM AUTHOR]

Gesture classification is one of the most emerging domains of today's era. Among the major applications of gesture classification, sign gesture recognition is considered the core class of this domain. Keeping in mind the specifications of the region and language-based variations, the development of a standard sign gesture classifier is the main objective of today's research. So far many language translators have been utilized to capture this communication deficiency. Software simulations and hardware prototype-based different models have been implemented using vision-based, sensor-based, and hybrid classifiers. This article focuses on the flex, accelerometer, and gyroscope sensors-based smart prototype which is designed to capture sign gestures. These sensors are placed on a glove to capture and compile digits i.e. 0 to 10, alphabets i.e. A–Z, and alphanumeric i.e. 0–10, A–Z datasets. The main objective of the proposed model is to classify sign gestures, made by deaf–mute individuals and recognize the actual meaning of performed gestures. K-nearest neighbor, discriminant analysis, and support vector machine-based machine learning algorithms are trained on our manually compiled datasets that are succeeded in achieving an average accuracy of 99.18%, 99.03%, and 99.82%, respectively. [ABSTRACT FROM AUTHOR]

In this work, a model-order reduction methodology based on proper orthogonal decomposition (POD) and Galërkin projection is presented and applied to the simulation of the self-ignition of a stockpile of solid fuel. Self-ignition is a phenomenon associated with steep changes in space and time, yielding high gradients of state variables which demand grid refinement and, thus, increase of the computational burden. To cope with this difficulty, first, a full order model (FOM), generated by finite-difference discretisation of the PDEs constituting the differential model, is employed to generate reference solutions. Two different POD-based formulations are proposed: the classical POD-Galërkin is employed to generate reduced order models (ROM), then discrete empirical interpolation method (DEIM) is employed to deal with nonlinearities in a more efficient manner. These reduction techniques are further supplemented with an innovative sampling approach based on k-means clustering. The resulting agile ROM is validated against the FOM. Both model-order reduction strategies, particularly the POD-DEIM model, reproduce the FOM solutions with high accuracy and much lower computational cost: The results of the application of a combination of the DEIM algorithm and k-means clustering show that the computational time for the calculation of one solution reduces up to 1020 times, while remaining able to reproduce all bifurcation points found with the FOM, thus demonstrating quantitative and qualitative agreement. [ABSTRACT FROM AUTHOR]

Saturation pressure is one the most important parameters required for describing a hydrocarbon reservoir. The determination of saturation pressure using conventional laboratory method is expensive and time consuming; hence, it is necessary to develop agile, easy and accurate saturation pressure prediction methods. The application of machine learning techniques for determination of saturation pressure for any fluid type has been shown in this study; no such work has been reported in this area so far. Four machine learning techniques namely K-Nearest Neighbor (KNN), Support vector machine (SVM), Light Gradient Boosting Machine (LGBM) and artificial neural network (ANN) were utilized. A universal novel correlation has also been developed using ANN. Models were developed utilizing experimentally determined composition of 305 reservoir oil and gas condensate, properties of heptane plus and reservoir temperature. Results concluded that models and proposed correlation predicted saturation pressure with high accuracy. Results indicated that the LGBM model has performed better with coefficient of determination (R2) and mean absolute error (MAE) values of 0.9655 and 194.3887, respectively, for complete data set. However, the ANN model was the most accurate model with R2 and MAE values of 0.9507 and 271.8008, respectively, for test data. [ABSTRACT FROM AUTHOR]

As governments access capital, technology, and managerial expertise from private investors, there is an increasing trend of privatizations in infrastructure projects worldwide. Given the large size of these investments, notably in the sector of telecommunications, investors typically create a consortium with other interested firms, an investment vehicle known as private participation project (PPP). Given the critical repercussions on the rest of the economy, and its large financial losses in case of failure, the prediction of the success of PPPs in telecommunications is of utmost importance. The success of PPPs can be predicted by Machine Learning and it is probed in this article. Hence, widely acknowledged classifiers (k-nearest neighbors [k-NNs], support vector machines [SVMs], and random forest [RF]) are applied to PPPs publicly available data from the World Bank. The results on this highly imbalanced dataset are greatly improved by the application of data balancing techniques. It includes some standard ones (random oversampling, random undersampling, and synthetic minority oversampling technique [SMOTE]), together with some other advanced ones (density-based SMOTE and borderline SMOTE). The satisfactory results validate the proposed application of classifiers on the dataset improved by data-balancing techniques. [ABSTRACT FROM AUTHOR]

In the present era of the medical field, the mortality rate of the infant increased by 1/3 percentage in a year. Even though we have modernized and have great expertise in the medical domain, we failed to control the infantile mortality rate. So continuous monitoring of the fetus during pregnancy is important. If there is any complication in the growth of the fetus, then the patient is put into the appropriate examination and medication proposed by the physician. In this paper, we propose methods to control the infant mortality rate in the early stage of pregnancy. Fetal heart rate of 2126 patients was collected and developed as datasets. Then with these datasets, we developed machine learning classifier models to classify the normal, suspect, and pathologic cases using the Decision tree, Naive Bayes, Random forest, and K-nearest neighbors. We divided the datasets into training dataset and testing dataset. Then the base classifier model was created using training datasets, and then the same models were verified by appending the test datasets. We improved the techniques and efficacy of the base classifiers by ensemble methods such as bagging and boosting. Finally, the datasets were classified as normal, suspect, and pathological cases with an improved accuracy of 96.617% with the help of the random forest classifier. [ABSTRACT FROM AUTHOR]

The Xin'an River ecological compensation mechanism is China's first horizontal compensation mechanism to improve the ecological environment of the downstream areas and increase the fiscal revenues of the upstream areas. Although the amount invested in the program is increasing every year, few studies have been conducted to evaluate the impact of the program. Based on the Levinsohn–Petrin method and a comprehensive dataset of Chinese industrial companies from 2009 to 2015, we employed the k-nearest neighbor propensity score matching–difference-in-difference (PSM-DID) to explore the impact of the Xin'an River Ecological Compensation Gambling Agreement (ECGA) on company technological progress and its drive mechanism. The empirical results show that ECGA inhibits the technological progress of upstream industrial companies and has more pronounced inhibitory effects for heavy-polluting companies. ECGA indirectly inhibits the technological progress in upstream companies by affecting profitability, scale, human capital, foreign direct investment, and management efficiency. [ABSTRACT FROM AUTHOR]

Preeclampsia (PE) presence could lead to hemodynamic changes. Previous research suggested that morphological parameters based on photoplethysmographic pulse waves (PPGW) could help diagnose PE. To investigate the performance of a novel PPGPW-based parameter, falling scaled slope (FSS), in distinguishing PE. To investigate the advantages of the machine learning algorithm over the conventional statistical methods in the analysis. Eighty-one pieces of PPGPW data were acquired for the study (PE, n = 44; normotensive, n = 37). The FSS values were calculated and used to construct a PE classifier using the K-nearest neighbors (KNN) algorithm. A predicted PE state varying from 0 to 1 was also calculated. The classifier's performance in distinguishing PE was evaluated using the ROC and AUC. A comparison was conducted with previously published PPGPW-based models. Compared to the previous PPGPW-based parameters, FSS showed a better performance in distinguishing PE with an AUC value of 0.924, the best threshold of 0.498 could predict PE with a sensitivity of 84.1% and a specificity of 89.2%. As for the analysis method, training a classifier using the KNN algorithm had an advantage over the conventional statistical methods with the AUC values of 0.878 and 0.749, respectively. The result indicated that FSS might be an effective tool for identifying PE. Moreover, the machine learning algorithm could further help the data analysis and improve performance. [ABSTRACT FROM AUTHOR]

It is very important to determine the parameters affecting the performance of the Darrieus-type wind turbine and its effects. In particular, it should be specified at which TSR value the peak power coefficient is obtained. In this study, standard and modified S1046 airfoils and aspect ratios (H/D), angle of attack (AoA), turbulent/non-turbulent flow (WT), number of blades (N), and chord length (C) were tested. Then, four different machines learning-based multi-output regression models (Decision Tree, Linear Regression, K-Nearest Neighbors, and Random Forest) were trained to make performance predictions with the data obtained from the evaluated test setup. Thirdly, feature selection based on the Random Forest algorithm, which is the best performing multi-output regression model, was performed using data due to changing parameter values on the established system. The importance of the parameters was determined. The operating range of the system was at relatively low TSR values. When analyzing the blade profile, the modified blade version performed better in certain combinations compared to the standard profile. Maximum power coefficient (Cp) was obtained from the modified turbine structure with 5 degrees of attack angle, H/D = 1.85, and C = 60 mm. The present study aims to increase the turbine's power coefficient and aims to predict results as power coefficient without doing many different experiments. [ABSTRACT FROM AUTHOR]

The parameters of thermal power, excess air, and supply air distance at burner pot are critical to obtain low emissions and high combustion efficiency in a pellet stove. The main aim of this study is to predict the performance and gaseous emissions in a pellet stove with four different supply airflow distances of 15, 30, 60, and 75 mm above the burner pot base at different excess air ratios and thermal power inputs. For this purpose, different models, namely, support vector regression, k-nearest neighbors, and deep neural network, are developed to predict the performance and emissions of the pellet stove. The thermal power, excess air ratio, and supply air distance were used as input variables to predict the flue gas temperature, efficiency, carbon monoxide (CO), nitrogen oxides (NOx), and carbon dioxide (CO2) emissions. The hyperparameters of the designed models are tuned using several optimization approaches: grid search, Bayesian optimization, particle swarm, and genetic algorithms. Experiments are conducted to validate the performances of the models with optimized hyperparameters using statistical metrics such as coefficient of determination, mean-squared error, mean absolute error, and mean absolute percentage error. Analysis results indicate that the deep neural network performs the best with the highest correlation coefficient and lowest error metrics in the prediction of output parameters. The coefficient of determination (R2) values of the DNN model were obtained as 0.8875, 0.8464, 0.934, 0.8692, and 0.9992 in the training phase 0.909, 0.8760, 0.9164, 0.9086, and 0.9991 in the testing phase for the efficiency, Tg, CO, NOx, and CO2 parameters, respectively. The results of this study can serve as a guide to help researchers, engineers, and facility owners or operators predict the gaseous emissions levels and to get an idea of the impact of operating conditions on performance and emissions behaviors. [ABSTRACT FROM AUTHOR]

This study, which aims to make predictions using a previously unused deep hybrid Convolutional Neural Network-Support Vector Regression approach for hydropower generation, was carried out using a unique meteorological data set as input data. The unique dataset was collected from Kaman Meteorology Directorate and Hirfanlı HEPP in Turkey, from 2007 to 2021, to estimate the HEPP's Net Head and Hydroelectric Power Generation on a daily basis. The performances of the prediction models were benchmarked in addition to the used CNN-SVR model with Machine Learning (ML) models (Boosting Random Forest Regression (BRFR) and Weighted K-Nearest Neighbor Regression (WKNNR)) and Deep Learning (DL) models (Long-Short Term Memory (LSTM) and Deep Belief Network (DBN)). The comparison results of the used CNN-SVR model with other alternative models showed that CNN-SVR performed effectively with the highest correlation coefficient of 0.971 for NH and 0.968 for PP. [ABSTRACT FROM AUTHOR]

Accurate land cover mapping is challenging in Southeast Asia where cloud coverage is prevalent and landscape is heterogenous. Object-based mapping, multi-temporal images and combined use of optical and microwave data, provide abundant features in spectral, spatial, temporal, geometric and polarimetric dimensions. And random forest is usually employed due to robustness and efficiency in handling high-dimensional and noisy data. This study assesses whether feature selection and ensemble analysis, which are rarely adopted, yield improved result. Recursive feature elimination decreases original 568 features into a subset of 53 features, achieving the optimal combination of features. Ensemble analysis of random forest, support vector machine, and K-nearest neighbors leads to an overall accuracy of 0.816. Comparison experiments demonstrated the merits of the multi-temporal, multi-source approach, feature elimination and ensemble analysis. [ABSTRACT FROM AUTHOR]

It is of great significance to identify soybean seeds with high oil content since the oil content of soybean seeds decides oil yield. At present, related researches mostly used machine learning algorithm to identify soybean varieties with small samples. In this study, 5800 spectral data samples of 58 varieties in the range of 400–1000 nm were obtained. An acceptable method that combines hyperspectral imaging with one-dimensional convolutional neural network was proposed to distinguish high oilcontent soybean seeds. Moreover, traditional machine learning models, including support vector machine, k-nearest neighbor algorithm, and partial squares discriminant analysis, were also established in the experimental study. The effects of four preprocessing methods, namely moving window smoothing, standard normal variate, multivariate scattering correction, and Savitzky–Golay, were compared when building support vector machine-based identification models. The results showed that the model using multivariate scattering correction gave better test accuracy (94.5%), indicating that for this study, multivariate scattering correction was a more suitable method than others. Meanwhile, the study compared the performance of the four models by expanding the number of samples. The results showed that the proposed one-dimensional convolutional neural network model was more stable. The average accuracy of the training set and test set was 96% and 93%, respectively. Therefore, hyperspectral data combined with one-dimensional convolutional neural network was effective in identifying soybean seeds with high oil content. [ABSTRACT FROM AUTHOR]

Honey quality is a global concern since this product is highly susceptible to adulteration, given its competitive price. As a reliable strategy for honey authenticity determination, this work introduces an intelligent classification system that considers the pattern recognition point of view to develop an economical and quick analytical method to identify and differentiate genuine from adulterated honey. This work used an electronic tongue composed of three working electrodes of carbon, platinum, and gold. The system used Cyclic voltammetry to obtain data from 50 genuine and 50 adulterated honey samples. Subsequently, the system used multivariate data analysis using a pattern recognition methodology composed of three big stages, including data organization and normalization, dimensionality reduction, and k-Nearest Neighbors (k-NN) as a classification method. The process was validated with the Leave One Out Cross Validation technique (LOOCV), reaching a classification accuracy performance of 100%. The results show that it was possible the development of a combined methodology between analytical tools and chemometrics for an in-situ, quick and efficient authenticity honey evaluation. [ABSTRACT FROM AUTHOR]

It is important to classify coal in the industry to improve its utilization. Herein, coal classification was performed using laser-induced breakdown spectroscopy (LIBS) combined with K-nearest neighbor (KNN) chemometrics. The principal component analysis was used to determine the optimum component of the original data. Eight elements (Al, Fe, Ca, Na, Mg, Si, Ti, and K) were selected as the indices for coal classification, while 11 elements were further divided into four categories as indicators for coal classification using the KNN model. The standard coal samples were divided based upon the ash and volatile values and the elemental content. The results were satisfactory, achieving an optimum accuracy of 97.73%. In contrast to traditional methods, LIBS significantly reduced the analysis time, simplified the process, and maintained high accuracy. [ABSTRACT FROM AUTHOR]

In recent years, the semiconductor industry is increasingly considering alternative computation schemes and in this conjuncture quantum computing has come up as a promising alternative due its ability to perform parallel computations (by exploiting superposition) and capacity to solve some of the intractable problems exponentially faster. But one of the criteria in the realization of quantum circuits is that corresponding descriptions demand a so-called nearest neighbour (NN) constraint to be satisfied. One of the possible solutions towards satisfying such constraints is through the insertion of SWAP gates that makes the qubit interactions within a gate adjacent. But since the incorporation of SWAP gates increases the cost of the circuits, strategies should be employed which satisfy the NN constraint while placing SWAP gates so that the cost are reduced. In this work, we propose improved design strategies to realize low-cost NN-compliant quantum circuits. Two approaches are introduced, where the first approach relies on a graph traversal process and uses heuristics in critical decision-making conditions to design a 1D NN-compliant representation. The second approach considers qubit-to-qubit interactions and realizes the circuits in a 2D platform based on some weighted metric. Both the designs are evaluated over a wide spectrum of benchmarks and results have been compared with related works. In those comparisons, we have seen that our approaches perform fairly better than the reported ones. [ABSTRACT FROM AUTHOR]

Landslide predictive performance is expected to vary with different sampling techniques, such as landslide random and cluster sampling. Current advancements in remote sensing technologies and machine learning (ML) have enhanced landslide prediction performance. The Himalayan Mountain range in Pakistan poses an unadorned threat to the ecosystem and valley population because of landslide occurrence. The present study explores, and tests alternative sampling technique based on spatial pattern characterization in the wake of increased landslide prediction efficacy, rather than a renowned random technique for training and testing sampling. Thereupon, landslide inventory data with 17 geo-environmental factors (i.e. topographic, hydrological and seismic factors) were determined. Landslide cluster patterns were confirmed by the Nearest Neighbor Index (NNI) method and after getting the cluster patterns, the predicted performance of landslide sampling was tested using ML and statistical methods. Advanced ML algorithms including Random Forest (RF), Extreme Gradient Boosting (XGBoost), Naive Bayes (NB), K-nearest Neighbors (KNN) and statistical methods including Weight-ofEvidence (WofE) and Logistic Regression (LR) were used and validated. The landslide-prone district of Azad Jammu and Kashmir (Neelum Valley), Kashmir Himalayas, Pakistan, was selected as a case study. Prediction performance rates are high with area under the curve (AUC) ranging from 0.802 to 0.912; accuracy (ACC) ranges from 0.78 to 0.89, and kappa ranges from 0.50 to 0.68 with cluster sampling technique, whereas the performance was low with random sampling technique, with AUC ranges from 0.768 to 0.895; ACC ranges from 0.74 to 0.86 and kappa ranges from 0.48 to 0.64. The descending order of accuracy of the six algorithms was XGboost, RF, KNN, NB, LR and WofE. Our results confirmed that the landslides followed cluster patterns in the study area, and ML algorithms with cluster training samples positively affected landslide susceptibility prediction with a statistically significant difference. The outcomes support the hypothesis that using landslides spatial natural existence, as training samples instead of random concepts, improves the prediction ability; and highlights that alternative landslide partitioning technique could be a practicable and robust choice for landslides prediction modelling. [ABSTRACT FROM AUTHOR]

Flooding is the main recurring natural disaster in Sungai Pinang catchment, Malaysia. Flash flood susceptibility mapping (FFSM) explains a key component of flood risk analysis and enables efficient estimation of the spatial extent of flood characteristics. The current study applied four machine learning models (i.e. Logistic Regression (LR), K-Nearest Neighbors (KNN), Random Forest (RF), and Extreme Gradient Boosting (XGBoost)) ensembled with the Statistical Index (SI) to develop flash flood susceptibility mapping (FFSM). 110 flash flood locations in the Sungai Pinang catchment were used in this study. Genetic algorithm (GA) was combined with Fuzzy Unordered Rules Induction Algorithm (FURIA), Rotation Forest, and Random Subspace for the feature selection method (FSM). The results showed that GA-FURIA outperformed the other two models in terms of accuracy based on the FSM. Twelve flash flood variables were selected by GA-FURIA. The FFSM results showed that the SI-RF model has the highest area under the receiver operating characteristics (AUROC) curve of success rate (0.978), whereas the SI-XGB has the best AUROC in terms of validation rate (0.997). The findings suggest that the twelve ideal conditioning variables may be used to optimize FFSM development. [ABSTRACT FROM AUTHOR]