Search

The glass relics are precious material evidence of the early trade and cultural exchange between the East and the West. To explore the cultural differences and trade development between early China and foreign countries, it is extremely important to classify glass cultural relics. Despite their similar appearances, Chinese glass contains more lead, while foreign glass contains more potassium. In view of this, this paper proposes a joint Daen-LR, ARIMA-LSTM, and MLR machine learning algorithm (JMLA) for the analysis and identification of the chemical composition of ancient glass. We separate the sampling points of ancient glass into two systems: lead-barium glass and high-potassium glass. Firstly, an improved logistic regression model based on a double adaptive elastic network (Daen-LR) is used to select variables with both Oracle and adaptive classification characteristics. Secondly, the ARIMA-LSTM model was used to establish the correlation curve of chemical composition before and after weathering and to predict the change in chemical composition with weathering. Thirdly, combining the data processed by the above two methods, a multiple linear regression model (MLR) is used to classify unknown glass products. It was shown that the sample obtained by this processing method has a very good fit. In comparison with other similar types of models like Decision Trees (DT), Random Forests (RF), Support Vector Machines (SVM), and Random Forests based on classification and regression trees (CART-RF), the classification accuracy of JMLA is 97.9% on the train set. The accuracy rate on the test set reached 97.6%. The results of the research demonstrate that JMLA can improve the accuracy of the glass type classification problem, greatly enhance the research efficiency of archaeological staff, and gain a more reliable result. [ABSTRACT FROM AUTHOR]

The acoustic cavitation of fluids, as well as related physical and chemical phenomena, causes a variety of effects that are highly important in technological processes and medicine. Therefore, it is important to be able to control the conditions that allow cavitation to begin and progress. However, the accurate prediction of acoustic cavitation is dependent on a complex relationship between external influence parameters and fluid characteristics. The multiparameter problem restricts the development of successful theoretical models. As a result, it is critical to identify the most important parameters influencing the onset of the cavitation process. In this paper, the ultrasonic frequency, hydrostatic pressure, temperature, degassing, density, viscosity, volume, and surface tension of a fluid were investigated using machine learning to determine their significance in predicting acoustic cavitation strength. Three machine learning models based on support vector regression (SVR), ridge regression (RR), and random forest (RF) algorithms with different input parameters were trained. The results showed that the SVM algorithm performed better than the other two algorithms. The parameters affecting the active cavitation nuclei, namely hydrostatic pressure, ultrasound frequency, and outgassing degree, were found to be the most important input parameters influencing the prediction of the cavitation threshold. Other parameters have a minor impact when compared to the first three, and their role can be compensated for by alternative variables. The further development of the obtained results provides a new way to optimize and improve existing theoretical models. [ABSTRACT FROM AUTHOR]

As a result of the continuous growth in the amount of geological data, machine learning (ML) offers an opportunity to contribute to solving problems in geosciences. However, digital geology applications introduce new challenges for machine learning due to the unique geoscience properties encountered in each problem, requiring novel research in ML. This paper proposes a novel machine learning method, entitled "Partial Decision Tree Forest (PART Forest)", to overcome these challenges introduced by geoscience problems and offers potential advancements in both machine learning and geoscience disciplines. The effectiveness of the proposed PART Forest method was illustrated in mineral classification. This study aims to build an intelligent ML model that automatically classifies the minerals in terms of their crystal structures (triclinic, monoclinic, orthorhombic, tetragonal, hexagonal, and trigonal) by taking into account their chemical compositions and their physical and optical properties. In the experiments, the proposed PART Forest method demonstrated its superiority over one of the well-known ensemble learning methods, random forest, in terms of accuracy, precision, recall, f-score, and AUC (area under the curve) metrics. [ABSTRACT FROM AUTHOR]

Flax fiber/shape memory epoxy hygromorph composites are a promising area of research in the field of biocomposites. This paper focuses on the tensile modulus of these composites and investigates how it is affected by factors such as fiber orientation (0° and 90°), temperature (20 °C, 40 °C, 60 °C, 80 °C, and 100 °C), and humidity (50% and fully immersed) conditions. Machine learning algorithms were utilized to predict the tensile modulus based on non-linearly dependent initial variables. Both decision tree (DT) and random forest (RF) algorithms were employed to analyze the data, and the results showed high coefficient of determination R2 values of 0.94 and 0.95, respectively. These findings demonstrate the effectiveness of machine learning in analyzing large datasets of mechanical properties in biocomposites. Moreover, the study revealed that the orientation of the flax fibers had the greatest impact on the tensile modulus value (with feature importance of 0.598 and 0.605 for the DT and RF models, respectively), indicating that it is a crucial factor to consider when designing these materials. [ABSTRACT FROM AUTHOR]

With the recent growth of the Internet, the volume of data has also increased. In particular, the increase in the amount of unstructured data makes it difficult to manage data. Classification is also needed in order to be able to use the data for various purposes. Since it is difficult to manually classify the ever-increasing volume data for the purpose of various types of analysis and evaluation, automatic classification methods are needed. In addition, the performance of imbalanced and multi-class classification is a challenging task. As the number of classes increases, so does the number of decision boundaries a learning algorithm has to solve. Therefore, in this paper, an improvement model is proposed using WordNet lexical ontology and BERT to perform deeper learning on the features of text, thereby improving the classification effect of the model. It was observed that classification success increased when using WordNet 11 general lexicographer files based on synthesis sets, syntactic categories, and logical groupings. WordNet was used for feature dimension reduction. In experimental studies, word embedding methods were used without dimension reduction. Afterwards, Random Forest (RF), Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) algorithms were employed to perform classification. These studies were then repeated with dimension reduction performed by WordNet. In addition to the machine learning model, experiments were also conducted with the pretrained BERT model with and without WordNet. The experimental results showed that, on an unstructured, seven-class, imbalanced dataset, the highest accuracy value of 93.77% was obtained when using our proposed model. [ABSTRACT FROM AUTHOR]

The paper presents the results of verification of the suitability of the random forest algorithm for the non-invasive assessment of excessively damp and salty historical brick walls. A new method of such quantitative assessment was developed and recently published by the author for the purpose of conducting research in buildings where destructive intervention is not possible due to conservation restrictions. However, before implementing the developed method into construction practice, it requires further validation. The conducted research showed that among all analyzed machine learning algorithms, the random forest algorithm is the most predisposed for the non-invasive evaluation of the Umc mass moisture content of brick walls. Data sets from archival research and experimental tests conducted in two historical buildings were used to verify the usefulness of this algorithm. This usefulness was confirmed by the obtained satisfactory values of the linear correlation coefficient R, which amounted to 0.801 for the first building and 0.803 for the second one. Moreover, it was also proved by the obtained low values of medians of the absolute errors |Δf| equal to 1.79% and 1.46%, and also by the not too high (for an in situ study) medians of the relative errors |RE| equal to 16.70% and 13.75%. [ABSTRACT FROM AUTHOR]

Urban municipal water supply is an important part of underground pipelines, and their scale continues to expand. Due to the continuous improvement in the quality and quantity of data available for pipeline systems in recent years, traditional pipeline network risk assessment cannot cope with the improvement of various monitoring methods. Therefore, this paper proposes a machine learning-based risk assessment method for municipal pipe network operation and maintenance and builds a model example based on the data of a pipeline network base in a park in Suzhou. We optimized the random forest learning model, compared it with other centralized learning methods, and finally evaluated the model's learning effect. Finally, the risk probability associated with each pipe segment sample was obtained, the risk factors affecting the pipe segment's failure were determined, and their relevance and importance ranking was established. The results showed that the most influential factors are pipe material, soil properties, service life, and the number of past failures. The random forest algorithm demonstrated better prediction accuracy and robustness on the dataset. [ABSTRACT FROM AUTHOR]

Multi-class classification is a fundamental task in Machine Learning. However, complex models can be viewed as black boxes, making it difficult to gain insight into how the model makes its predictions and build trust in its decision-making process. This paper presents a novel method called Multi-Class Fuzzy-LORE (mcFuzzy-LORE) for explaining the decisions made by multi-class fuzzy-based classifiers such as Fuzzy Random Forests (FRF). mcFuzzy-LORE is an adaptation of the Fuzzy-LORE method that uses fuzzy decision trees as an alternative to classical decision trees, providing interpretable, human-readable rules that describe the reasoning behind the model's decision for a specific input. The proposed method was evaluated on a private dataset that was used to train an FRF-based multi-class classifier that assesses the risk of developing diabetic retinopathy in diabetic patients. The results show that mcFuzzy-LORE outperforms prior classical LORE-based methods in the generation of counterfactual instances. [ABSTRACT FROM AUTHOR]

Simple Summary: Pig face recognition plays an important role in the intelligent breeding and accurate management of pigs, and the mobile and embedded applications of this technology in the management of small and medium-sized pig farms are in great demand; therefore, in order to make the model more suitable for use in small and medium-sized pig farms, in this study, PCA pre-treatment was added to the traditional method. The experiment shows that the model is suitable for small and medium-sized pig farms, and it can promote the intelligent transformation of pig breeding management. To explore the application of a traditional machine learning model in the intelligent management of pigs, in this paper, the influence of PCA pre-treatment on pig face identification with RF is studied. By this testing method, the parameters of two testing schemes, one adopting RF alone and the other adopting RF + PCA, were determined to be 65 and 70, respectively. With individual identification tests carried out on 10 pigs, accuracy, recall, and f1-score were increased by 2.66, 2.76, and 2.81 percentage points, respectively. Except for the slight increase in training time, the test time was reduced to 75% of the old scheme, and the efficiency of the optimized scheme was greatly improved. It indicates that PCA pre-treatment positively improved the efficiency of individual pig identification with RF. Furthermore, it provides experimental support for the mobile terminals and the embedded application of RF classifiers. [ABSTRACT FROM AUTHOR]

In this paper we predict Bitcoin movements by utilizing a machine-learning framework. We compile a dataset of 24 potential explanatory variables that are often employed in the finance literature. Using daily data from 2nd of December 2014 to July 8th 2019, we build forecasting models that utilize past Bitcoin values, other cryptocurrencies, exchange rates and other macroeconomic variables. Our empirical results suggest that the traditional logistic regression model outperforms the linear support vector machine and the random forest algorithm, reaching an accuracy of 66%. Moreover, based on the results, we provide evidence that points to the rejection of weak form efficiency in the Bitcoin market. [ABSTRACT FROM AUTHOR]

This paper presents a methodology to estimate the multidimensional poverty index using spatial data at the street block level. The data used in this study were obtained from Open Street Maps and ESA's land use cover, which are freely available sources of spatial information. The study employs five machine-learning algorithms, including Catboost, Lightboost, and Random Forest, to estimate the multidimensional poverty index with spatial granularity. The results indicate that these models achieve promising performance in predicting poverty levels in Medellín, Colombia. The results showed that the Random Forest algorithm achieved the highest performance, with an MAE of 0.07504. Furthermore, the spatial distribution of the multidimensional poverty estimate was highly correlated with the true values of the distribution. This work contributes to predicting multidimensional poverty by demonstrating the potential of machine learning algorithms to utilize accessible spatial data. By providing evidence of the feasibility of estimating poverty levels at a granular spatial level, this methodology offers a powerful tool for policymakers to make poverty social interventions with low-cost evidence. Furthermore, this study has important implications for poverty eradication efforts in developing countries, where access to reliable data remains challenging. [ABSTRACT FROM AUTHOR]

At present, the condition assessment of transformer winding based on frequency response analysis (FRA) measurements demands skilled personnel. Despite many research efforts in the last decade, there is still no definitive methodology for the interpretation and condition assessment of transformer winding based on FRA results, and this is a major challenge for the industrial application of the FRA method. To overcome this challenge, this paper proposes a transformer condition assessment (TCA) algorithm, which is based on numerical indices, and a supervised machine learning technique to develop a method for the automatic interpretation of FRA results. For this purpose, random forest (RF) classifiers were developed for the first time to identify the condition of transformer winding and classify different faults in the transformer windings. Mainly, six common states of the transformer were classified in this research, i.e., healthy transformer, healthy transformer with saturated core, mechanically damaged winding, short-circuited winding, open-circuited winding, and repeatability issues. In this research, the data from 139 FRA measurements performed in more than 80 power transformers were used. The database belongs to the transformers having different ratings, sizes, designs, and manufacturers. The results reveal that the proposed TCA algorithm can effectively assess the transformer winding condition with up to 93% accuracy without much human intervention. [ABSTRACT FROM AUTHOR]

Split Hopkinson pressure bar (SHPB) tests are usually used to determine the dynamic mechanical strength of basalt-fiber-reinforced concrete (BFRC), but this test method is time-consuming and expensive. This paper makes predictions about the dynamic mechanical strength of BFRC by employing machine learning (ML) algorithms and feature sets drawn from experimental data from prior works. However, there is still the problem of improving the accuracy of the dynamic mechanical strength prediction by the BFRC, which remains a challenge. Using stacking ensemble learning and genetic algorithms (GA) to optimize parameters, this study proposes a prediction method that combines these two techniques for obtaining accurate predictions. This method is composed of three parts: (1) the training uses multiple base learners, and the algorithms employed by the learners include extreme gradient boosting (XGBoost), gradient boosting (GB), random forest (RF), and support vector regression (SVR); (2) multi-base learners are combined using a stacking strategy to obtain the final prediction; and (3) using GA, the parameters are optimized in the prediction model. An experiment was conducted to compare the proposed approach with popular techniques for machine learning. In the study, the stacking ensemble algorithm integrated the base learner prediction results to improve the model's performance and the GA further improved prediction accuracy. As a result of the application of the method, the dynamic mechanical strength of BFRC can be predicted with high accuracy. A SHAP analysis was also conducted using the stacking model to determine how important the contributing properties are and the sensitivity of the stacking model. Based on the results of this study, it was found that in the SHPB test, the strain rate had the most significant influence on the DIF, followed by the specimen diameter and the compressive strength. [ABSTRACT FROM AUTHOR]

In this paper, we propose a radio environment map (REM) update methodology based on clustering and machine learning for indoor coverage. We use real measurements collected by the TurtleBot3 mobile robot using the received signal strength indicator (RSSI) as a measure of link quality between transmitter and receiver. We propose a practical framework for timely updates to the REM for dynamic wireless communication environments where we need to deal with variations in physical element distributions, environmental factors, movements of people and devices, and so on. In the proposed approach, we first rely on a historical dataset from the area of interest, which is used to determine the number of clusters via the K-means algorithm. Next, we divide the samples from the historical dataset into clusters, and we train one random forest (RF) model with the corresponding historical data from each cluster. Then, when new data measurements are collected, these new samples are assigned to one cluster for a timely update of the RF model. Simulation results validate the superior performance of the proposed scheme, compared with several well-known ML algorithms and a baseline scheme without clustering. [ABSTRACT FROM AUTHOR]

Landslide susceptibility assessment (LSA) based on machine learning methods has been widely used in landslide geological hazard management and research. However, the problem of sample imbalance in landslide susceptibility assessment, where landslide samples tend to be much smaller than non-landslide samples, is often overlooked. This problem is often one of the important factors affecting the performance of landslide susceptibility models. In this paper, we take the Wanzhou district of Chongqing city as an example, where the total number of data sets is more than 580,000 and the ratio of positive to negative samples is 1:19. We oversample or undersample the unbalanced landslide samples to make them balanced, and then compare the performance of machine learning models with different sampling strategies. Three classic machine learning algorithms, logistic regression, random forest and LightGBM, are used for LSA modeling. The results show that the model trained directly using the unbalanced sample dataset performs the worst, showing an extremely low recall rate, indicating that its predictive ability for landslide samples is extremely low and cannot be applied in practice. Compared with the original dataset, the sample set optimized through certain methods has demonstrated improved predictive performance across various classifiers, manifested in the improvement of AUC value and recall rate. The best model was the random forest model using over-sampling (O_RF) (AUC = 0.932). [ABSTRACT FROM AUTHOR]

The latest international video coding standard, H.266/Versatile Video Coding (VVC), supports high-definition videos, with resolutions from 4 K to 8 K or even larger. It offers a higher compression ratio than its predecessor, H.265/High Efficiency Video Coding (HEVC). In addition to the quadtree partition structure of H.265/HEVC, the nested multi-type tree (MTT) structure of H.266/VVC provides more diverse splits through binary and ternary trees. It also includes many new coding tools, which tremendously increases the encoding complexity. This paper proposes a fast intra coding algorithm for H.266/VVC based on visual perception analysis. The algorithm applies the factor of average background luminance for just-noticeable-distortion to identify the visually distinguishable (VD) pixels within a coding unit (CU). We propose calculating the variances of the numbers of VD pixels in various MTT splits of a CU. Intra sub-partitions and matrix weighted intra prediction are turned off conditionally based on the variance of the four variances for MTT splits and a thresholding criterion. The fast horizontal/vertical splitting decisions for binary and ternary trees are proposed by utilizing random forest classifiers of machine learning techniques, which use the information of VD pixels and the quantization parameter. Experimental results show that the proposed algorithm achieves around 47.26% encoding time reduction with a Bjøntegaard Delta Bitrate (BDBR) of 1.535% on average under the All Intra configuration. Overall, this algorithm can significantly speed up H.266/VVC intra coding and outperform previous studies. [ABSTRACT FROM AUTHOR]

Pipeline corrosion is one of the leading causes of failures in the transmission of gas and hazardous liquids in the oil and gas industry. In-line inspection is a non-destructive inspection for detecting corrosion defects in pipelines. Defects are measured in terms of their width, length and depth. Consecutive in-line inspection data are used to determine the pipeline's corrosion growth rate and its remnant life, which set the operational and maintenance activities of the pipeline. The traditional approach of manually processing in-line inspection data has various weaknesses, including being time consuming due to huge data volume and complexity, prone to error, subject to biased judgement by experts and challenging for matching of in-line inspection datasets. This paper aimed to contribute to the adoption of machine learning approaches in classifying pipeline defects as per Pipeline Operator Forum requirements and matching in-line inspection data for determining the corrosion growth rate and remnant life of pipelines. Machine learning techniques, namely, decision tree, random forest, support vector machines and logistic regression, were applied in the classification of pipeline defects using Phyton programming. The performance of each technique in terms of the accuracy of results was compared. The results showed that the decision tree classifier model was the most accurate (99.9%) compared with the other classifiers. [ABSTRACT FROM AUTHOR]

A cerebral stroke is a medical problem that occurs when the blood flowing to a section of the brain is suddenly cut off, causing damage to the brain. Brain cells gradually die because of interruptions in blood supply and other nutrients to the brain, resulting in disabilities, depending on the affected region. Early recognition and detection of symptoms can aid in the rapid treatment of strokes and result in better health by reducing the severity of a stroke episode. In this paper, the Random Forest (RF), Extreme Gradient Boosting (XGBoost), and light gradient-boosting machine (LightGBM) were used as machine learning (ML) algorithms for predicting the likelihood of a cerebral stroke by applying an open-access stroke prediction dataset. The stroke prediction dataset was pre-processed by handling missing values using the KNN imputer technique, eliminating outliers, applying the one-hot encoding method, and normalizing the features with different ranges of values. After data splitting, synthetic minority oversampling (SMO) was applied to balance the stroke samples and no-stroke classes. Furthermore, to fine-tune the hyper-parameters of the ML algorithm, we employed a random search technique that could achieve the best parameter values. After applying the tuning process, we stacked the parameters to a tuning ensemble RXLM that was analyzed and compared with traditional classifiers. The performance metrics after tuning the hyper-parameters achieved promising results with all ML algorithms. [ABSTRACT FROM AUTHOR]

The Quick UDP Internet Connections (QUIC) protocol provides advantages over traditional TCP, but its encryption functionality reduces the visibility for operators into network traffic. Many studies deploy machine learning and deep learning algorithms on QUIC traffic classification. However, standalone machine learning models are subject to overfitting and poor predictability in complex network traffic environments. Deep learning on the other hand requires a huge dataset and intensive parameter fine-tuning. On the contrary, ensemble techniques provide reliability, better prediction, and robustness of the trained model, thereby reducing the chance of overfitting. In this paper, we approach the QUIC network traffic classification problem by utilizing five different ensemble machine learning techniques, namely: Random Forest, Extra Trees, Gradient Boosting Tree, Extreme Gradient Boosting Tree, and Light Gradient Boosting Model. We used the publicly available dataset with five different services such as Google Drive, YouTube, Google Docs, Google Search, and Google Music. The models were trained using a different number of features on different scenarios and evaluated using several performance metrics. The results show that Extreme Gradient Boosting Tree and Light Gradient Boosting Model outperform the other models and achieve one of the highest results among the state-of-the-art models found in the literature with a simpler model and features. [ABSTRACT FROM AUTHOR]

This paper examines applying machine learning to the assessment of the quality of the transmission in optical networks. The motivation for research into this problem derives from the fact that the accurate assessment of transmission quality is key to an effective management of an optical network by a network operator. In order to facilitate a potential implementation of the proposed solution by a network operator, the training data for the machine learning algorithms are directly extracted from an operating network via a control plane. Particularly, this work focuses on the application of single class and binary classification machine learning algorithms to optical network transmission quality assessment. The results obtained show that the best performance can be achieved using gradient boosting and random forest algorithms. [ABSTRACT FROM AUTHOR]

Phishing is an online threat where an attacker impersonates an authentic and trustworthy organization to obtain sensitive information from a victim. One example of such is trolling, which has long been considered a problem. However, recent advances in phishing detection, such as machine learning-based methods, have assisted in combatting these attacks. Therefore, this paper develops and compares four models for investigating the efficiency of using machine learning to detect phishing domains. It also compares the most accurate model of the four with existing solutions in the literature. These models were developed using artificial neural networks (ANNs), support vector machines (SVMs), decision trees (DTs), and random forest (RF) techniques. Moreover, the uniform resource locator's (URL's) UCI phishing domains dataset is used as a benchmark to evaluate the models. Our findings show that the model based on the random forest technique is the most accurate of the other four techniques and outperforms other solutions in the literature. [ABSTRACT FROM AUTHOR]

With the application and development of Internet technology, network traffic is growing rapidly, and the situation of network security is becoming more and more serious. As an important way to protect network security, abnormal traffic detection has been paid more and more attention. In this paper, the uncertainty of the samples in the abnormal traffic detection dataset is studied. Combining the three-way decision idea with the random forest algorithm, a three-way selection random forest optimization model for abnormal traffic detection is proposed. Firstly, the three-way decision idea is integrated into the random selection process of feature attributes, and the attribute importance based on decision boundary entropy is calculated. The feature attributes are divided into the normal domain, abnormal domain, and uncertain domain, and the three-way attribute random selection rules are designed to randomly select the feature attributes that conform to the rules from different domains. Secondly, the classifier evaluation function is constructed by combining pure accuracy and diversity, and the anomaly traffic detection base classifier with a high evaluation value is selected for integration to eliminate the unstable factors caused by randomness in the process of base classifier generation. Thirdly, the optimal node weight combination of the base classifier is obtained by iterative calculation of the gray wolf optimization algorithm to further improve the prediction effect and robustness of the model. Finally, the model is applied to the abnormal traffic detection dataset. The experimental results show that the prediction accuracy of the three-way selection random forest optimization model on CIC-IDS2017, KDDCUP99, and NSLKDD datasets is 96.1%, 95.2%, and 95.3%, respectively, which has a better detection effect than other machine learning algorithms. [ABSTRACT FROM AUTHOR]

This paper proposes a non-parametric technique of option valuation and hedging. Here, we replicate the extended Black–Scholes pricing model for the exotic barrier options and their corresponding Greeks using the fully connected feed-forward neural network. Our methodology involves some benchmarking experiments, which result in an optimal neural network hyperparameter that effectively prices the barrier options and facilitates their option Greeks extraction. We compare the results from the optimal NN model to those produced by other machine learning models, such as the random forest and the polynomial regression; the output highlights the accuracy and the efficiency of our proposed methodology in this option pricing problem. The results equally show that the artificial neural network can effectively and accurately learn the extended Black–Scholes model from a given simulated dataset, and this concept can similarly be applied in the valuation of complex financial derivatives without analytical solutions. [ABSTRACT FROM AUTHOR]

Most of the current, popular approaches to monitoring real driving NOx emissions are based on direct measurement. However, due to the uncertainty of sensor-based measurements, such methods cannot always be used to accurately screen out the malfunctions of an emission control system. In this paper, a random forest (RF) model which extracts information from on-board diagnostics (OBD) data streams transmitted by a remote emission management vehicle terminal (REMVT) is proposed to provide a specific emission method for the online screening of high NOx emissions. First, two particular forms of modeling, random forest and logistic regression (LR), are laid out as representatives of nonparametric models and specified linear models. These two models were trained, validated and compared using OBD data collected from three China-VI heavy-duty diesel vehicles (HDDVs). The results show that as a data-driven, highly adaptive and robust learning method, the RF model can more accurately identify an abnormal emission state. Finally, a further validation was conducted, in which another China-VI HDDV was tested in two typical states, including a fault state and a normal state. The results indicated that the RF model could clearly distinguish the out-of-control emission condition from the normal operation state. The outcome of this research verifies the feasibility of using a machine learning model to process remote OBD data on HD vehicles and to identify high emissions in the case of an in-use fleet. On this basis, more sophisticated combined models and multi-stage models could be developed. [ABSTRACT FROM AUTHOR]

Neurological diseases are a significant health threat, often presenting through abnormalities in electroencephalogram (EEG) signals during seizures. In recent years, machine learning (ML) technologies have been explored as a means of automated EEG pathology diagnosis. However, existing ML-based EEG binary classification methods largely focus on extracting EEG-related features, which may lead to poor performance in classifying EEG signals by overlooking potentially redundant information. In this paper, we propose a novel Kruskal–Wallis (KW) test-based framework for EEG pathology detection. Our framework first divides EEG data into frequency sub-bands using wavelet packet decomposition and then extracts statistical characteristics from each selected coefficient. Next, the piecewise aggregation approximation technique is used to obtain the aggregated feature vectors, followed by the KW statistical test methodology to select significant features. Finally, three ensemble learning classifiers, random forest, categorical boosting (CatBoost), and light gradient boosting machine, are used to classify the extracted significant features into normal or abnormal classes. Our proposed framework achieves an accuracy of 89.13%, F1-score of 87.60%, and G-mean of 88.60%, respectively, outperforming other competing techniques on the same dataset, which shows the great promise in EEG pathology detection. [ABSTRACT FROM AUTHOR]

Accurate solar forecasting lately relies on advances in the field of artificial intelligence and on the availability of databases with large amounts of information on meteorological variables. In this paper, we present the methodology applied to introduce a large-scale, public, and solar irradiance dataset, CyL-GHI, containing refined data from 37 stations found within the Spanish region of Castile and León (Spanish: Castilla y León, or CyL). In addition to the data cleaning steps, the procedure also features steps that enable the addition of meteorological and geographical variables that complement the value of the initial data. The proposed dataset, resulting from applying the processing methodology, is delivered both in raw format and with the quality processing applied, and continuously covers 18 years (the period from 1 January 2002 to 31 December 2019), with a temporal resolution of 30 min. CyL-GHI can result in great importance in studies focused on the spatial-temporal characteristics of solar irradiance data, due to the geographical information considered that enables a regional analysis of the phenomena (the 37 stations cover a land area larger than 94,226 km2). Afterwards, three popular artificial intelligence algorithms were optimised and tested on CyL-GHI, their performance values being offered as baselines to compare other forecasting implementations. Furthermore, the ERA5 values corresponding to the studied area were analysed and compared with performance values delivered by the trained models. The inclusion of previous observations of neighbours as input to an optimised Random Forest model (applying a spatio-temporal approach) improved the predictive capability of the machine learning models by almost 3%. Dataset:  https://doi.org/10.5281/zenodo.7404167 Dataset License: CC-BY-SA-NC [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) has become widely adopted in businesses, organizations, and daily lives. They are usually characterized by transferring and processing sensitive data. Attackers have exploited this prospect of IoT devices to compromise user data's integrity and confidentiality. Considering the dynamic nature of the attacks, artificial intelligence (AI)-based techniques incorporating machine learning (ML) are promising techniques for identifying such attacks. However, the dataset being utilized features engineering techniques, and the kind of classifiers play significant roles in how accurate AI-based predictions are. Therefore, for the IoT environment, there is a need to contribute more to this context by evaluating different AI-based techniques on datasets that effectively capture the environment's properties. In this paper, we evaluated various ML models with the consideration of both binary and multiclass classification models validated on a new dedicated IoT dataset. Moreover, we investigated the impact of different features engineering techniques including correlation analysis and information gain. The experimental work conducted on bagging, k-nearest neighbor (KNN), J48, random forest (RF), logistic regression (LR), and multi-layer perceptron (MLP) models revealed that RF achieved the highest performance across all experiment sets, with a receiver operating characteristic (ROC) of 99.9%. [ABSTRACT FROM AUTHOR]

A simple supervised learning model can predict a class from trained data based on the previous learning process. Trust in such a model can be gained through evaluation measures that ensure fewer misclassification errors in prediction results for different classes. This can be applied to supervised learning using a well-trained dataset that covers different data points and has no imbalance issues. This task is challenging when it integrates a semi-supervised learning approach with a dynamic data stream, such as social network data. In this paper, we propose a stream-based evolving bot detection (SEBD) framework for Twitter that uses a deep graph neural network. Our SEBD framework was designed based on multi-view graph attention networks using fellowship links and profile features. It integrates Apache Kafka to enable the Twitter API stream and predict the account type after processing. We used a probably approximately correct (PAC) learning framework to evaluate SEBD's results. Our objective was to maintain the accuracy and confidence levels of our framework to enable successful learning with low misclassification errors. We assessed our framework results via cross-domain evaluation using test holdout, machine learning classifiers, benchmark data, and a baseline tool. The overall results show that SEBD is able to successfully identify bot accounts in a stream-based manner. Using holdout and cross-validation with a random forest classifier, SEBD achieved an accuracy score of 0.97 and an AUC score of 0.98. Our results indicate that bot accounts participate highly in hashtags on Twitter. [ABSTRACT FROM AUTHOR]

The main research question addressed in the paper is related to the possibility of medium-term forecasting of the public debts of the EU member states. The analysis focuses on a broad range of indicators (macroeconomic, fiscal, monetary, global, and convergence) that influence the public debt levels of the EU member states. A machine learning prediction model using random forest regression was approbated with the empirical data. The algorithm was applied in two iterations—a primary iteration with 33 indicators and a secondary iteration with the 8 most significant indicators in terms of their influence and forecasting importance regarding the development of public debt across the EU. The research identifies a change in the medium term (2023–2024) in the group of the four most indebted EU member states, viz., that Spain will be replaced by France, which is an even more systemic economy, and will thus increase the group's share of the EU's GDP. The results indicate a logical scenario of rising interest rates with adverse effects for the fiscal imbalances, which will require serious reforms in the public sector of the most indebted EU member states. [ABSTRACT FROM AUTHOR]

The importance of civil engineering infrastructure in modern societies has increased lately due to the growth of the global economy. It forges global supply chains facilitating enormous economic activity. The bridges usually form critical links in complex supply chain networks. Structural health monitoring (SHM) of these infrastructures is essential to reduce life-cycle costs, and determine their remaining life using advanced sensing techniques and data fusion methods. However, the data obtained from the SHM systems describing the health condition of the infrastructure systems may contain anomalies (i.e., distortion, drift, bias, outlier, noise etc.). An automated framework is required to accurately classify these anomalies and evaluate the current condition of these systems in a timely and cost-effective manner. In this paper, a recursive and interpretable decision tree framework is proposed to perform multiclass classification of acceleration data collected from a real-life bridge. The decision nodes of the decision tree are random forest classifiers that are invoked recursively after synthetically augmenting the training data before successive iterations until suitable classification performance is obtained. This machine-learning-based classification model evolved from a simplistic decision tree where statistical features are used to perform classification. The feature vectors defined for training the random forest classifiers are calculated using similar statistical features that are easy to interpret, enhancing the interpretability of the classifier models. The proposed framework could classify non-anomalous (i.e., normal) time-series of the test dataset with 98% accuracy. [ABSTRACT FROM AUTHOR]

Analyzing the impact of climate variables into the operational planning processes is essential for the robust implementation of a sustainable power system. This paper deals with the modeling of the run-of-river hydropower production based on climate variables on the European scale. A better understanding of future run-of-river generation patterns has important implications for power systems with increasing shares of solar and wind power. Run-of-river plants are less intermittent than solar or wind but also less dispatchable than dams with storage capacity. However, translating time series of climate data (precipitation and air temperature) into time series of runof-river-based hydropower generation is not an easy task as it is necessary to capture the complex relationship between the availability of water and the generation of electricity. This task is also more complex when performed for a large interconnected area. In this work, a model is built for several European countries by using machine learning techniques. In particular, we compare the accuracy of models based on the Random Forest algorithm and show that a more accurate model is obtained when a finer spatial resolution of climate data is introduced. We then discuss the practical applicability of a machine learning model for the medium term forecasts and show that some very context specific but influential events are hard to capture. [ABSTRACT FROM AUTHOR]

Online frauds have pernicious impacts on different system domains, including waste management systems. Fraudsters illegally obtain rewards for their recycling activities or avoid penalties for those who are required to recycle their own waste. Although some approaches have been introduced to prevent such fraudulent activities, the fraudsters continuously seek new ways to commit illegal actions. Machine learning technology has shown significant and impressive results in identifying new online fraud patterns in different system domains such as e-commerce, insurance, and banking. The purpose of this paper, therefore, is to analyze a waste management system and develop a machine learning model to detect fraud in the system. The intended system allows consumers, individuals, and organizations to track, monitor, and update their performance in their recycling activities. The data set provided by a waste management organization is used for the analysis and the model training. This data set contains transactions of users’ recycling activities and behaviors. Three machine learning algorithms, random forest, support vector machine, and multi-layer perceptron are used in the experiments and the best detection model is selected based on the model’s performance. Results show that each of these algorithms can be used for fraud detection in waste managements with high accuracy. The random forest algorithm produces the optimal model with an accuracy of 96.33%, F1-score of 95.20%, and ROC of 98.92%. [ABSTRACT FROM AUTHOR]

This document is an editorial for a special issue of the journal Micromachines on the topic of exploring IoT sensors and their applications in smart environments. The editor highlights the importance of IoT sensors in various fields and discusses the advancements and challenges in this area. The editorial introduces ten research contributions that cover a range of applications and technologies, including wearable health monitoring systems, gesture recognition devices, low-light pedestrian detection systems, and gas sensors. The editor emphasizes the need for interdisciplinary collaboration, scalability of sensors, and data security in future research. Overall, this special issue provides valuable insights into the current state of IoT sensors and sets the stage for future innovations in this field. [Extracted from the article]

Digital twin (DT) is a key technology for realizing the interconnection and intelligent operation of the physical world and the world of information and provides a new paradigm for fault diagnosis. Traditional machine learning algorithms require a balanced dataset. Training and testing sets must have the same distribution. Training a good generalization model is difficult in an actual production line operation process. Fault diagnosis technology based on the digital twin uses its ultrarealistic, multisystem, and high-precision characteristics to simulate fault data that are difficult to obtain in an actual production line to train a reliable fault diagnosis model. In this article, we first propose an improved random forest (IRF) algorithm, which reselects decision trees with high accuracy and large differences through hierarchical clustering and gives them weights. Digital twin technology is used to simulate a large number of balanced datasets to train the model, and the trained model can be transferred to a physical production line through transfer learning for fault diagnosis. Finally, the feasibility of our proposed algorithm is verified through a case study of an automobile rear axle assembly line, for which the accuracy of the proposed algorithm reaches 97.8%. The traditional machine learning plus digital twin fault diagnosis method proposed in this paper involves some generalization, and thus has practical value when extended to other fields. [ABSTRACT FROM AUTHOR]

Rice has long served as the staple food in Asia, and the cultivation of high-yield rice crops draws increasing attention from academic researchers. The prediction of rice growth condition by image features realizes nondestructive prediction and it has great implications for smart agriculture. We found a special image parameter called the fractal dimension that can improve the effect of the prediction model. As an important geometric feature, the fractal dimension could be calculated from the image, but it is rarely used in the field of rice growth prediction. In this paper, we attempt to combine the fractal dimension with traditional rice image features to improve the effect of the model. The thresholding method is used to transform the cropped rice image into binary image, and the box-counting method is used to calculate the fractal dimension of the image. The correlation coefficients are calculated to select the characteristics with a strong correlation with biomass. The prediction models of dry weight, fresh weight and plant height of rice are established by using random forest, support vector regression and linear regression. By evaluating the prediction effect of the model, it can be concluded that the fractal dimension can improve the prediction effect of the model. Among the models obtained by the three methods, the multiple linear regression model has the best comprehensive effect, with the dry weight prediction model R 2 reaching 0.8697, the fresh weight prediction model R 2 reaching 0.8631 and the plant height prediction model R 2 reaching 0.9196. The model established in this paper has a fine effect and has a certain guiding significance in rice research. [ABSTRACT FROM AUTHOR]

The planning and management of traffic flow networks with multiple input data sources for decision-making generate the need for a mathematical approach. The program of measures for the development of the transport infrastructure of the Russian Federation provides for the selection of pilot regions for the creation of intelligent transportation systems. With extensive knowledge of theoretical and applied mathematics, it is important to select and adapt mathematical methods for solving problems. In this regard, the aim of the study is to develop and validate an algorithm for solving the problem of classifying objects according to the potential of creating intelligent transportation systems. The main mathematical apparatus for classification is the «random forest» machine learning algorithm method. A bagging machine learning meta-algorithm for high accuracy of the algorithm was used. This paper proposes the author's method of sequential classification analysis for identifying objects with the potential to create intelligent transportation systems. The choice of using this method is justified by its best behavior under the large number of predictor variables required for an objective aggregate assessment of digital development and quality of territories. The proposed algorithm on the example of Russian regions was tested. A technique and algorithm for statistical data processing based on descriptive analytics tools have been developed. The quality of the classification analysis algorithm was assessed by the random forest method based on misclassification coefficients. The admissibility of retrained algorithms and formation of a «fine-grained» «random forest» model for solving classification problems under the condition of no prediction was proven to be successful. The most productive models with the highest probability of correct classification were «reached» and «finalized» on the basis of logistic regression analysis of relationships between predictors and categorical dependent variables. The regions of class 1 with «high potential for the creation of intelligent transportation systems» are most likely to be ready for the reorganization of infrastructure facilities; the introduction of digital technologies in the management of traffic flows was found. [ABSTRACT FROM AUTHOR]

Orphan genes (OGs) may evolve from noncoding sequences or be derived from older coding material. Some shares of OGs are present in all sequenced genomes, participating in the biochemical and physiological pathways of many species, while many of them may be associated with the response to environmental stresses and species-specific traits or regulatory patterns. However, identifying OGs is a laborious and time-consuming task. This paper presents an automated predictor, XGBoost-A2OGs (identification of OGs for angiosperm based on XGBoost), used to identify OGs for seven angiosperm species based on hybrid features and XGBoost. The precision and accuracy of the proposed model based on fivefold cross-validation and independent testing reached 0.90 and 0.91, respectively, outperforming other classifiers in cross-species validation via other models, namely, Random Forest, AdaBoost, GBDT, and SVM. Furthermore, by analyzing and subdividing the hybrid features into five sets, it was proven that different hybrid feature sets influenced the prediction performance of OGs involving eudicot and monocot groups. Finally, testing of small-scale empirical datasets of each species separately based on optimal hybrid features revealed that the proposed model performed better for eudicot groups than for monocot groups. [ABSTRACT FROM AUTHOR]

Plate end (PE) debonding and intermediate crack (IC) debonding are the two main failure modes of beams strengthened with fiber-reinforced polymer (FRP) in flexure. Therefore, it is essential to clarify the force state of the structure when debonding occurs in strengthened beams. This paper collected 229 beams with debonding failure as the database, of which 128 were PE debonding and 101 were IC debonding. Correlation and grey correlation analysis were used to establish the indicator systems for predicting PE and IC debonding and to identify the critical indicators among them. Five machine learning models, linear regression, ridge regression, decision trees, random forests, and back propagation (BP) neural networks, were used to build the two debonding prediction models. Optimization of the best prediction among the five machine learning models took place using the Dung Beetle Optimizer (DBO) algorithm, which has competitive performance with state-of-the-art optimization approaches in terms of convergence rate, solution accuracy, and stability. Finally, the optimal prediction model was compared with the models suggested by codes, and it was found that the established model can well predict PE and IC debonding. [ABSTRACT FROM AUTHOR]

As the most widely used storage device today, hard disks are efficient and convenient, but the damage incurred in the event of a failure can be very significant. Therefore, early warnings before hard disk failure, allowing the stored content to be backed up and transferred in advance, can reduce many losses. In recent years, an endless stream of research on the prediction of hard disk failure prediction has emerged. The detection accuracy of various methods, from basic machine learning models, such as decision trees and random forests, to deep learning methods, such as BP neural networks and recurrent neural networks, has also been improving. In this paper, based on the idea of blending ensemble learning, a novel failure prediction method combining machine learning algorithms and neural networks is proposed on the publicly available BackBlaze hard disk datasets. The failure prediction experiment is conducted only with S.M.A.R.T., that is, the learned characteristics collected by self-monitoring analysis and reporting technology, which are internally counted during the operation of the hard disk. The experimental results show that this ensemble learning model is able to outperform other independent models in terms of evaluation criterion based on the Matthews correlation coefficient. Additionally, through the experimental results on multiple types of hard disks, an ensemble learning model with high performance on most types of hard disks is found, which solves the problem of the low robustness and generalization of traditional machine learning methods and proves the effectiveness and high universality of this method. [ABSTRACT FROM AUTHOR]

It is obvious that one of the most significant challenges posed by Twitter is the proliferation of fraudulent and fake accounts, as well as the challenge of identifying these accounts. As a result, the primary focus of this paper is on the identification of fraudulent accounts, fake information, and fake accounts on Twitter, in addition to the flow of content that these accounts post. The research utilized a design science methodological approach and developed a bot account referred to as "Fake Account Detector" that assists with the detection of inappropriate posts that are associated with fake accounts. To develop this detector, previously published tweets serve as the datasets for the training session. This data comes from Twitter and are obtained through the REST API. The technique of machine learning with random forest (RF) is then used to train the data. The high levels of accuracy (99.4%) obtained from the RF detection results served as the foundation for the development of the bot account. This detector tool, developed using this model, can be utilized by individuals, businesses, and government agencies to assist in the detection and prevention of Twitter problems related to fake news and fake accounts. [ABSTRACT FROM AUTHOR]

In software development, the main problem is recognizing the security-oriented issues within the reported bugs due to their unacceptable failure rate to provide satisfactory reliability on customer and software datasets. The misclassification of bug reports has a direct impact on the effectiveness of the bug prediction model. The misclassification issue surely compromises the accuracy of the system. Manually reviewing bug reports is necessary to solve this problem, but doing so takes a lot of time and is tiresome for developers and testers. This paper proposes a novel hybrid approach based on natural language processing (NLP) and machine learning. To address these issues, the intended outcomes are multi-class supervised classification and bug prioritization using supervised classifiers. After being collected, the dataset was prepared for vectorization, subjected to exploratory data analysis, and preprocessed. The feature extraction and selection methods used for a bag of words are TF-IDF and word2vec. Machine learning models are created after the dataset has undergone a full transformation. This study proposes, develops, and assesses four classifiers: multinomial Naive Bayes, decision tree, logistic regression, and random forest. The hyper-parameters of the models are tuned, and it is concluded that random forest outperformed with a 91.73% test and 100% training accuracy. The SMOTE technique was used to balance the highly imbalanced dataset, which was initially created for the justified classification. The comparison between balanced and imbalanced dataset models clearly showed the importance of the balanced dataset in classification as it outperformed in all experiments. [ABSTRACT FROM AUTHOR]

The transmission of information, ideas, and thoughts requires communication, which is a crucial component of human contact. The utilization of Internet of Things (IoT) devices is a result of the advent of enormous volumes of messages delivered over the internet. The IoT botnet assault, which attempts to perform genuine, lucrative, and effective cybercrimes, is one of the most critical IoT dangers. To identify and prevent botnet assaults on connected computers, this study uses both quantitative and qualitative approaches. This study employs three basic machine learning (ML) techniques—random forest (RF), decision tree (DT), and generalized linear model (GLM)—and a stacking ensemble model to detect botnets in computer network traffic. The results reveled that random forest attained the best performance with a coefficient of determination (R2) of 0.9977, followed by decision tree with an R2 of 0.9882, while GLM was the worst among the basic machine learning models with an R2 of 0.9522. Almost all ML models achieved satisfactory performance, with an R2 above 0.93. Overall, the stacking ensemble model obtained the best performance, with a root mean square error (RMSE) of 0.0084 m, a mean absolute error (MAE) of 0.0641 m, and an R2 of 0.9997. Regarding the stacking ensemble model as compared with the single machine learning models, the R2 of the stacking ensemble machine learning increased by 0.2% compared to the RF, 1.15% compared to the DT, and 3.75% compared to the GLM, while RMSE decreased by approximately 0.15% compared to the GLM, DT, and RF single machine learning techniques. Furthermore, this paper suggests best practices for preventing botnet attacks. Businesses should make major investments to combat botnets. This work contributes to knowledge by presenting a novel method for detecting botnet assaults using an artificial-intelligence-powered solution with real-time behavioral analysis. This study can assist companies, organizations, and government bodies in making informed decisions for a safer network that will increase productivity. [ABSTRACT FROM AUTHOR]

Logistic regression is a very popular binary classification technique in many industries, particularly in the financial service industry. It has been used to build credit scorecards, estimate the probability of default or churn, identify the next best product in marketing, and many more applications. The machine learning literature has recently introduced several alternative techniques, such as deep learning neural networks, random forests, and factorisation machines. While neural networks and random forests form part of the practitioner's model-building toolkit, factorisation machines are seldom used. In this paper, we investigate the applicability of factorisation machines to some binary classification problems in banking. To stimulate the practical application of factorisation machines, we implement the fitting routines, based on logit loss and maximum likelihood, on commercially available software that is widely used by banks and other large financial services companies. Logit loss is usually used by the machine learning fraternity while maximum likelihood is popular in statistics. Depending on the coding of the target variable, we will show that these methods yield identical parameter estimates. Often, banks are confronted with predicting events that occur with low probability. To deal with this phenomenon, we introduce weights in the above-mentioned loss functions. The accuracy of our fitting algorithms is then studied by means of a simulation study and compared with logistic regression. The separation and prediction performance of factorisation machines are then compared to logistic regression and random forests by means of three case studies covering a recommender system, credit card fraud, and a credit scoring application. We conclude that logistic factorisation machines are worthy competitors of logistic regression in most applications, but with clear advantages in recommender systems applications where the number of predictors typically outnumbers the number of observations. [ABSTRACT FROM AUTHOR]

Recent years have witnessed booming data on drugs and their associated adverse drug reactions (ADRs). It was reported that these ADRs have resulted in a high hospitalisation rate worldwide. Therefore, a tremendous amount of research has been carried out to predict ADRs in the early phases of drug development, with the goal of reducing possible future risks. The pre-clinical and clinical phases of drug research can be time-consuming and cost-ineffective, so academics are looking forward to more extensive data mining and machine learning methods to be applied in this field of study. In this paper, we try to construct a drug-to-drug network based on non-clinical data sources. The network presents underlying relationships between drug pairs according to their common ADRs. Then, multiple node-level and graph-level network features are extracted from this network, e.g., weighted degree centrality, weighted PageRanks, etc. After concatenating the network features to the original drug features, they were fed into seven machine learning models, e.g., logistic regression, random forest, support vector machine, etc., and were compared to the baseline, where there were no network-based features considered. These experiments indicate that all the tested machine-learning methods would benefit from adding these network features. Among all these models, logistic regression (LR) had the highest mean AUROC score (82.1%) across all ADRs tested. Weighted degree centrality and weighted PageRanks were identified to be the most critical network features in the LR classifier. These pieces of evidence strongly indicate that the network approach can be vital in future ADR prediction, and this network-based approach could also be applied to other health informatics datasets. [ABSTRACT FROM AUTHOR]

Cryptojacking is a type of computer piracy in which a hacker uses a victim's computer resources, without their knowledge or consent, to mine for cryptocurrency. This is made possible by new memory-based cryptomining techniques and the growth of new web technologies such as WebAssembly, allowing mining to occur within a browser. Most of the research in the field of cryptojacking has focused on detection methods rather than prevention methods. Some of the detection methods proposed in the literature include using static and dynamic features of in-browser cryptojacking malware, along with machine learning algorithms such as Support Vector Machine (SVM), Random Forest (RF), and others. However, these methods can be effective in detecting known cryptojacking malware, but they may not be able to detect new or unknown variants. The existing prevention methods are shown to be effective only against web-assembly (WASM)-based cryptojacking malware and cannot handle mining service-providing scripts that use non-WASM modules. This paper proposes a novel hybrid approach for detecting and preventing web-based cryptojacking. The proposed approach performs the real-time detection and prevention of in-browser cryptojacking malware, using the blacklisting technique and statistical code analysis to identify unique features of non-WASM cryptojacking malware. The experimental results show positive performances in the ease of use and efficiency, with the detection accuracy improved from 97% to 99.6%. Moreover, the time required to prevent already known malware in real time can be decreased by 99.8%. [ABSTRACT FROM AUTHOR]

The prediction of drug-target interactions (DTIs) is a vital step in drug discovery. The success of machine learning and deep learning methods in accurately predicting DTIs plays a huge role in drug discovery. However, when dealing with learning algorithms, the datasets used are usually highly dimensional and extremely imbalanced. To solve this issue, the dataset must be resampled accordingly. In this paper, we have compared several data resampling techniques to overcome class imbalance in machine learning methods as well as to study the effectiveness of deep learning methods in overcoming class imbalance in DTI prediction in terms of binary classification using ten (10) cancer-related activity classes from BindingDB. It is found that the use of Random Undersampling (RUS) in predicting DTIs severely affects the performance of a model, especially when the dataset is highly imbalanced, thus, rendering RUS unreliable. It is also found that SVM-SMOTE can be used as a go-to resampling method when paired with the Random Forest and Gaussian Naïve Bayes classifiers, whereby a high F1 score is recorded for all activity classes that are severely and moderately imbalanced. Additionally, the deep learning method called Multilayer Perceptron recorded high F1 scores for all activity classes even when no resampling method was applied. [ABSTRACT FROM AUTHOR]

Distributed Denial of Service (DDoS) is still one of the main threats to network security today. Attackers are able to run DDoS in simple steps and with high efficiency to slow down or block users' access to services. In this paper, we propose a framework based on feature and model selection (FAMS), which is used for detecting DDoS attacks with the aim of identifying the features and models with a high generalization capability, high prediction accuracy, and short prediction time. The FAMS framework is divided into four main phases. The first phase is data pre-processing, including operations such as feature coding, outlier processing, duplicate elimination, data balancing, and normalization. In the second stage, 79 features are extracted from the dataset and selected by the feature selection algorithms filter, wrapper, embedded, variance, mutual information, backward elimination, Lasso.L1, and random forest. The purpose of feature selection is to simplify the model, avoid dimensional disasters, reduce computational costs, and reduce the prediction time. The third stage is model selection, which aims to select the most ideal algorithm from GD, SVM, LR, RF, HVG, SVG, HVR, and SVR using a model selection algorithm for the selected 21 features, and the results show that RF is far ahead in all evaluation indexes compared to the other models. The fourth stage is model optimization, which aims to further improve the performance of the RF algorithm in detecting DDoS attacks by optimizing the parameters max_samples, max_depth, n_estimators for the initially selected RF by the RF optimization algorithm. Finally, by testing the 100,000 CIC-IDS2018, CIC-IDS2017, and CIC-DoS2016 synthetic datasets, the results show that all the results have achieved excellent performance in the same category. Moreover, the framework also shows an excellent generalization performance by testing over 1 million synthetic datasets and over 330,000 CIC-DDoS2019 datasets. [ABSTRACT FROM AUTHOR]

Considering that driving stress is a major contributor to traffic accidents, detecting drivers' stress levels in time is helpful for ensuring driving safety. This paper attempts to investigate the ability of ultra-short-term (30-s, 1-min, 2-min, and 3-min) HRV analysis for driver stress detection under real driving circumstances. Specifically, the t-test was used to investigate whether there were significant differences in HRV features under different stress levels. Ultra-short-term HRV features were compared with the corresponding short-term (5-min) features during low-stress and high-stress phases by the Spearman rank correlation and Bland–Altman plots analysis. Furthermore, four different machine-learning classifiers, including a support vector machine (SVM), random forests (RFs), K-nearest neighbor (KNN), and Adaboost, were evaluated for stress detection. The results show that the HRV features extracted from ultra-short-term epochs were able to detect binary drivers' stress levels accurately. In particular, although the capability of HRV features in detecting driver stress also varied between different ultra-short-term epochs, MeanNN, SDNN, NN20, and MeanHR were selected as valid surrogates of short-term features for driver stress detection across the different epochs. For drivers' stress levels classification, the best performance was achieved with the SVM classifier, with an accuracy of 85.3% using 3-min HRV features. This study makes a contribution to building a robust and effective stress detection system using ultra-short-term HRV features under actual driving environments. [ABSTRACT FROM AUTHOR]

The diagnosis and prognosis of cardiovascular disease are crucial medical tasks to ensure correct classification, which helps cardiologists provide proper treatment to the patient. Machine learning applications in the medical niche have increased as they can recognize patterns from data. Using machine learning to classify cardiovascular disease occurrence can help diagnosticians reduce misdiagnosis. This research develops a model that can correctly predict cardiovascular diseases to reduce the fatality caused by cardiovascular diseases. This paper proposes a method of k-modes clustering with Huang starting that can improve classification accuracy. Models such as random forest (RF), decision tree classifier (DT), multilayer perceptron (MP), and XGBoost (XGB) are used. GridSearchCV was used to hypertune the parameters of the applied model to optimize the result. The proposed model is applied to a real-world dataset of 70,000 instances from Kaggle. Models were trained on data that were split in 80:20 and achieved accuracy as follows: decision tree: 86.37% (with cross-validation) and 86.53% (without cross-validation), XGBoost: 86.87% (with cross-validation) and 87.02% (without cross-validation), random forest: 87.05% (with cross-validation) and 86.92% (without cross-validation), multilayer perceptron: 87.28% (with cross-validation) and 86.94% (without cross-validation). The proposed models have AUC (area under the curve) values: decision tree: 0.94, XGBoost: 0.95, random forest: 0.95, multilayer perceptron: 0.95. The conclusion drawn from this underlying research is that multilayer perceptron with cross-validation has outperformed all other algorithms in terms of accuracy. It achieved the highest accuracy of 87.28%. [ABSTRACT FROM AUTHOR]

In this paper, the growing significance of data analysis in manufacturing environments is exemplified through a review of relevant literature and a generic framework to aid the ease of adoption of regression-based supervised learning in manufacturing environments. To validate the practicality of the framework, several regression learning techniques are applied to an open-source multi-stage continuous-flow manufacturing process data set to typify inference-driven decision-making that informs the selection of regression learning methods for adoption in real-world manufacturing environments. The investigated regression learning techniques are evaluated in terms of their training time, prediction speed, predictive accuracy (R-squared value), and mean squared error. In terms of training time ( T T ), k-NN20 (k-Nearest Neighbour with 20 neighbors) ranks first with average and median values of 4.8 ms and 4.9 ms, and 4.2 ms and 4.3 ms, respectively, for the first stage and second stage of the predictive modeling of the multi-stage continuous-flow manufacturing process, respectively, over 50 independent runs. In terms of prediction speed ( P S ), DTR (decision tree regressor) ranks first with average and median values of 5.6784 × 10 6 observations per second (ob/s) and 4.8691 × 10 6 observations per second (ob/s), and 4.9929 × 10 6 observations per second (ob/s) and 5.8806 × 10 6 observations per second (ob/s), respectively, for the first stage and second stage of the predictive modeling of the multi-stage continuous-flow manufacturing process, respectively, over 50 independent runs. In terms of R-squared value ( R 2 ), BR (bagging regressor) ranks first with average and median values of 0.728 and 0.728, respectively, over 50 independent runs, for the first stage of the predictive modeling of the multi-stage continuous-flow manufacturing process, and RFR (random forest regressor) ranks first with average and median values of 0.746 and 0.746, respectively, over 50 independent runs, for the second stage of the predictive modeling of the multi-stage continuous-flow manufacturing process. In terms of mean squared error ( M S E ), BR (bagging regressor) ranks first with average and median values of 2.7 and 2.7, respectively, over 50 independent runs, for the first stage of the predictive modeling of the multi-stage continuous-flow manufacturing process, and RFR (random forest regressor) ranks first with average and median values of 3.5 and 3.5, respectively, over 50 independent runs, for the second stage of the predictive modeling of the multi-stage continuous-flow manufacturing process. All methods are further ranked inferentially using the statistics of their performance metrics to identify the best method(s) for the first and second stages of the predictive modeling of the multi-stage continuous-flow manufacturing process. A Wilcoxon rank sum test is then used to statistically verify the inference-based rankings. D T R and k-NN20 have been identified as the most suitable regression learning techniques given the multi-stage continuous-flow manufacturing process data used for experimentation. [ABSTRACT FROM AUTHOR]