Search

In light of the extensive contamination of water sources by microplastics, their substantial specific surface area makes them favorable candidates as adsorbents for the simultaneous removal of coexisting contaminants in wastewater. In this regard, polyethylene microplastics were utilized to eliminate methylene blue dye from water. MB adsorption onto microplastics reached equilibrium in just 30 min at pH 7. The better fit of fractional power and Redlich-Peterson models on kinetic and equilibrium adsorption data, respectively, revealed that the MB removal process is a chemisorption in multilayer adsorption on the heterogeneous surface of the microplastics particles. The reusability of the microplastics adsorbent was confirmed based on the promising outcomes observed after five cycles. The results of the random forest regression exhibited an R2 of 97.55% for the correlation between the model-computed and measured amounts of MB reduction. The sensitivity analysis illustrated that the MB sorption process on microplastics is highly influenced by the initial MB concentration and adsorbent mass. These results show that although microplastics may pose potential risks to water environments, their adsorption potential can be utilized to simultaneously omit other pollutants from the aqueous solutions. [ABSTRACT FROM AUTHOR]

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

Rock physics plays an important role in the oil, gas, and water industries by providing essential data for reservoir management. This study focuses on predicting Shear Wave Velocity (Vs), a key geophysical parameter, using the Deep Belief Network (DBN) and Random Forest (RF) algorithms. The dataset is divided for training, validation, and testing, revealing that the DBN model surpasses the RF and empirical models in accuracy. With a low RMSE value of 0.0398 km/s in testing and an R2 of 0.9775, the DBN model demonstrates precision in Vs prediction. The analysis identifies specific parameters—Measure Depth (MD), Caliper (CALL), Gamma Ray (GR), Shallow Electrical Resistivity (RES-S), Medium Electrical Resistivity (RES-M), Deep Electrical Resistivity (RES-D), Compressional Wave Velocity (DCT), Neutron Porosity (NPHI), and Photoelectric Coefficient Index (PEF)—as significantly influencing Vs, while CALI and GR have a relatively lesser effect. The results of the Vs prediction indicated that the DBN algorithm excels in deep feature extraction, unsupervised pretraining, and efficient handling of complex data. Ultimately, the DBN algorithm proves to be a superior and time-saving alternative for Vs prediction and analysis in subsurface reservoirs. • Rock physics aids oil, gas, and water industries through crucial reservoir data. • Deep Belief Network (DBN) excels in Vs prediction with superior performance. • DBN model achieves remarkable precision with a low RMSE of 0.0398 km/s. • Correlation analysis identifies key parameters influencing shear wave velocity (Vs). • DBN algorithm proves a time-saving, reliable alternative for Vs prediction in subsurface reservoirs. [ABSTRACT FROM AUTHOR]

Landslides present a significant hazard to human life, infrastructure, and property, particularly in mountainous regions. In Morocco, these risks have garnered increased attention due to their detrimental impact. This study seeks to model landslide susceptibility using three machine learning classifiers (MLCs): Multi-Layer Perceptron (MLP), Random Forest (RF), and Adaptive Boosting Classifier (AdaBoost), and compare their performance. Initially, 144 landslide sites were identified, and thirteen factors pertaining to landslides were considered. The models' performance was assessed by calculating the area under the receiver operating characteristic curve (AUC-ROC). The findings reveal that AUC values range from 68.7% for AdaBoost to 82.2% for RF. The generated landslide susceptibility maps can aid decision-makers in avoiding areas with a high susceptibility to landslides. [ABSTRACT FROM AUTHOR]

This study addresses the critical issue of classifying cardiac ischemia, a disease with signifi- cant global health implications that contributes to the global mortality rate. In our study, we tackle the classification of ischemia using six diverse electrocardiogram (ECG) datasets and a convolutional neural network (CNN) as the primary methodology. We combined six separate datasets to gain a more comprehensive understanding of cardiac electrical activity, utilizing 12 leads to obtain a broader perspective. A discrete wavelet transform (DWT) preprocessing was used to eliminate irrelevant information from the signals, aiming to improve classification results. Focusing on accuracy and minimizing false negatives (FN) in ischemia detection, we enhance our study by incorporating various machine learning models into our base model. These models include multilayer perceptron (MLP), support vector machines (SVM), random forest (RF), long short-term memory (LSTM), and bidirectional LSTM (BiLSTM), allowing us to leverage the strengths of each algorithm. The CNN-BiLSTM model achieved the highest accuracy of 99.23% and demonstrated good sensitivity of 98.53%, effectively reducing false negative cases in the overall tests. The CNN-BiLSTM model demonstrated the ability to effectively identify abnormalities, misclassifying only 25 out of 1,673 ischemic cases in the test set as normal. This is due to the BiLSTM's efficiency in capturing long-range dependencies and sequential patterns, making it suitable for tasks involving time-series data such as ECG signals. In addition, CNNs are well-suited for hierarchical feature learning and complex pattern recognition in ECG data. [ABSTRACT FROM AUTHOR]

Web attacks often target web applications because they can be accessed over a network and often have vulnerabilities. The success of an intrusion detection system (IDS) in detecting web attacks depends on an effective traffic classification system. Several previous studies have utilized machine learning classification methods to create an efficient IDS with various datasets for different types of attacks. This paper utilizes the Canadian Institute for Cyber Security's (CIC-IDS2017) IDS dataset to assess web attacks. Importantly, the dataset contains 80 attributes of recent assaults, as reported in the 2016 McAfee report. Three machine learning algorithms have been evaluated in this research, namely random forests (RF), k-nearest neighbor (KNN), and naive bayes (NB). The primary goal of this research is to propose an effective machine learning algorithm for the IDS web attacks model. The evaluation compares the performance of three algorithms (RF, KNN, and NB) based on their accuracy and precision in detecting anomalous traffic. The results indicate that the RF outperformed the NB and KNN in terms of average accuracy achieved during the training phase. During the testing phase, the KNN algorithm outperformed others, achieving an average accuracy of 99.4916%. However, RF and KNN achieved 100% average precision and recall rates compared to other algorithms. Finally, the RF and KNN algorithms have been identified as the most effective for detecting IDS web attacks. [ABSTRACT FROM AUTHOR]

Climate information can be obtained from General circulation models (GCMs). However, this model has poor resolution, so it is necessary to do bias correction to overcome this problem. This study carried out a bias correction process using the Support Vector Regression (SVR) and Random Forest (RF) approaches. Bias correction is carried out for temperature in Indonesia using the BNU-ESM and MERRA-2 climate models, which act as observational data. The results show that the RF method (RMSE: 0.334; Correlation: 0.694; Standard Deviation: 0.582) is better than SVR (RMSE: 0.341; Correlation: 0.675; Standard Deviation: 0.588) in performing bias correction. [ABSTRACT FROM AUTHOR]

Predicting the property insurance claim has become very important due to the increasing demand for property and the fluctuating number of monthly claims. However, current prediction methods suffer from limited feature important information and imbalanced claim data. This research introduces a machine learning approach by considering demographic factors of insured objects and marketing strategies to predict claim trends. The study also tackles data imbalance using the SMOTE technique. Results reveal that random forest outperforms logistic regression, achieving a recall value of 96%. The research findings enable insurance companies to better analyze insured risk profiles and manage the number of claims. [ABSTRACT FROM AUTHOR]

Spare-parts provisioning strategy is crucial for airlines or Maintenance Repair Overhauls (MRO). The strategy is divided into multiple classes that refer to several criteria, then uses Material Requirements Planning (MRP) to run those strategies. However, classifying spare parts into a strategy is onerous and requires innovation. Therefore, we employ three algorithms: Decision Tree, Random Forest, and Support Vector Machine (SVM), and use spare parts and aircraft utilization data to classify the strategy and optimize spare parts provisioning of MRO over time. The results showed Random Forest performs best by accuracy, sensitivity, and specificity with a score of more than 97%. [ABSTRACT FROM AUTHOR]

Currently, in the process of assessing and giving feedback on students' argumentative writing, educators have to spend a considerable amount of time reading and analyzing each essay individually. This can be a complicated and time-consuming process, especially if the number of students to be assessed is quite large. The problem of this research is to find the most effective algorithm in providing accurate and reliable predictions in the context of evaluation and feedback of students' argumentation. This study compares three algorithms (logistic regression, Naive Bayes, and Random Forest) to predict student argumentation using essays from grades 6-12. Logistic regression performed best with 94.34% accuracy, followed by random forest with 91.98% accuracy, and Naive Bayes with 88.93% accuracy. The study optimized preprocessing and selected algorithms for an automated guidance model. It is the first stage of a three-part study for developing automated guidance models. Data came from Kaggle, and the study aims to improve the accuracy of automated guidance models for student argumentation. [ABSTRACT FROM AUTHOR]

In the present era, everyone is connected via the Internet for sharing digital information. The digital data is stored using the cloud technology. However, cloud technology is speedily increasing the volume of digital information and network intrusions. In this case, safeguarding the cloud data is essential for several purposes. Therefore, the present study emphasizes developing the network intrusion detection system using the benchmark NSL-KDD datasets. The ensemble learning-enabled random forest algorithm was proposed and implemented to select the most suitable features. The network intrusion detection and classification have been done using three machine learning models: support vector machine (SVM), logistic regression, and K-nearest neighbour's (KNN) with 87.58, 88.86, and 98.24% validation accuracies. Thus, the present study approach can be used in real-time cyber-attack detection and monitoring. [ABSTRACT FROM AUTHOR]

The overwhelming burden on the healthcare system has resulted in increased mortality rates as individuals struggle to receive appropriate medical attention. Machine learning has demonstrated its efficacy across diverse applications, fueling a growing trend in the realm of innovative automation. This is particularly facilitated by drug recommendation systems, which possess the capability to significantly reduce the dependence on specialized assistance. Faced with limited access to medical expertise, a considerable number of individuals have turned to self-medication, often worsening their health conditions. In response to this critical scenario, this paper introduces an inventive drug recommendation system which helps to lessen the strain healthcare professionals. Here, we present the development of a medication recommendation system based on LSTM, utilizing user-provided symptoms and the model's evaluations of past drug recommendations. The results of the comparison between Random Forest, SVM, and LSTM indicated that the modified LSTM-based recommendation system outperforms the other two models, achieving a 94% accuracy rate. [ABSTRACT FROM AUTHOR]

The leading cause of mortality in the world for the past few decades has been cardiovascular disease (CVD) or cardiac arrest, which is caused by conditions that affect the Heart. With Machine Learning (ML), predictive models are refined and adapted to test health industries in a more efficient way. ML algorithms will helpful to make an early prediction for those who may need to make a change to their lifestyle and receive appropriate medical treatment. Our research presents supervised ML classifiers capable of detecting significant features using machine learning techniques to predict Heart disease. An over sampling method is presented here for learning from unbalanced datasets. Several classification techniques are employed with different features in our prediction model. Performance metrics are analyzed to determine their effectiveness when applied to the UCI heart dataset using ML techniques by SMOTE oversampling; A maximum of 96.6 percentage of accuracy, 90 percentage of sensitivity, and 100 percentage of specificity and precision were achieved by the Random Forest method. [ABSTRACT FROM AUTHOR]

Missing values are a prevalent issue, frequently leading to a considerable decline in the quality of data. Therefore, it becomes imperative to adeptly manage missing data. This study presents a technique for imputing missing values that incorporates Kernelized Fuzzy C-Means (KFCM) clustering. The technique introduced is termed RFKFCM, which combines the benefits of Random Forest (RF) and KFCM algorithms. The proposed RFKFCM's performance is evaluated through a comparative analysis with five state-of-the-art imputation techniques (mean, LI, KNN, FCM, and LIFCM) across four widely used real-world datasets obtained from the UCI repository. Additionally, experiments investigating the impact of missing values affirm the robustness of the proposed technique across different missing rates using Friedman's statistical test for each dataset by comparing their mean rank. The experimental results indicate that our proposed technique performs significantly better than the existing imputation techniques based on Root mean squared error (RMSE) & Mean absolute error (MAE) for these datasets. [ABSTRACT FROM AUTHOR]

Given the increasing number of countries reporting degraded air quality, effective air quality monitoring has become a critical issue in today's world. However, the current air quality observatory systems are often prohibitively expensive, resulting in a lack of observatories in many regions within a country. Consequently, a significant problem arises where not every region receives the same level of air quality information. This disparity occurs because some locations have to rely on information from observatories located far away from their regions, even if they may be the closest available options. To address this challenge, a novel approach that leverages machine learning and deep learning techniques to forecast fine dust concentrations was proposed. Specifically, continuous location features in the form of latitude and longitude values were incorporated into our models. By utilizing a comprehensive dataset comprising weather conditions, air quality measurements, and location properties, various machine learning models, including Random Forest Regression, XGBoost Regression, AdaBoost Regression, and a deep learning model known as Long Short-Term Memory (LSTM) were trained. Our experimental results demonstrated that the LSTM model outperforms the other models, achieving the best score with a root mean squared error of 23.48 in predicting fine dust (PM10) concentrations on an hourly basis. Furthermore, the fact that incorporating location properties, such as longitude and latitude values, enhances the overall quality of the regression models was discovered. Additionally, the implications and contributions of our research were discussed. By implementing our approach, the cost associated with relying solely on existing observatories can be substantially reduced. This reduction in costs can pave the way for economically efficient fine dust observation systems, ensuring more widespread and accurate air quality monitoring across different regions. [ABSTRACT FROM AUTHOR]

Vehicles become an inevitable factor in everyone's life. Sometimes it becomes a threat to human lives and society. For any real-time-based applications, everyone should focus on predicting failure-prone components. A vehicle's air pressure system (APS) is one of its most important parts. If any system failure happens against APS it leads to core-financial losses, which in turn sometimes leads to loss of human lives. Prediction of APS negligence in a real-time application requires a deep diagnosis and diligent solution. In this study, we developed a machine learning model to predict system failure against APS. A real-time dataset that includes the 170 features and the presence of high-class imbalance data and missing values has been taken and experimentally validated with existing linear and nonlinear classifiers. The performance metrics results show that the Random Forest classifier exceeds other algorithms for training and testing data with an accuracy and F1 score of 99.5 and 99.5 percent respectively. [ABSTRACT FROM AUTHOR]

With the advancements in computer networking, communication between end-to-end systems has increased drastically. However, security issues have also been raised. Thus, detecting anomalies from a complex cloud environment is still challenging. Therefore, the present article proposes the deep Convolutional Neural Network (CNN) model for detecting and classifying near-real-time network intrusions from an imbalanced cloud environment. The random forest model is also offered and implemented to select the best suitable features as input to the CNN model. The experiments were carried out on CSE-CIC-IDS2018 datasets. The results show that the proposed CNN model achieved 97.07% testing accuracy with a 2.93% error rate. The performance of the proposed model was also measured using precision, recall, and f1-score with 98.11, 96.93, and 97.52%. The results are more accurate, precise, promising, and able to detect network anomalies with the highest accuracy and can be successfully used in real-time Industry 4.0 systems. [ABSTRACT FROM AUTHOR]

In conventional machining processes based on the chip removal mechanism, the progressive wear of the tool determines a change of the geometric characteristics of the cutting edge. Tool wear is a complex phenomenon and related tool life depends on several factors, such as cutting parameters, lubrication, and tool-workpiece relative trajectories. Tool wear progression affects the quality of the machined parts, making the tool replacement necessary even before its breakage. Moreover, in industrial practice, tool replacement cannot depend on the subjectivity of the operator, thus, the definition of an optimized strategy for cutting tool wear monitoring before tool failure is mandatory. This work compares Random Forest and Neural Network models for predicting tool wear in drilling. For the development of predictive models, tool life tests were performed by drilling through holes on AISI 9840 steel parts, with a coated tungsten carbide drill of 8 mm of diameter and by using constant cutting parameters. Flank wear of the tool was monitored. A set of statistical features computed from the vibration signals, the acoustic emission signals, the power signals and the torque signals constitute the input of the algorithms. The classification accuracy was 88% and 91% for Neural Network and Random Forest models respectively. In order to correctly define the tool replacement policy during production, the developed Random Forest model has been implemented in an industry, through production management software, achieving promising results. [ABSTRACT FROM AUTHOR]

Inaccurate cost estimates significantly affect the ultimate cost of construction projects and reduce revenues. This study compares the accuracy of two algorithm models in estimating construction project prices during the early conceptual stage of project development. The construction prices of residential projects are forecasted in this study using two modeling methods, artificial neural networks (ANN) and random forest (RF), based on the project data from the actual projects. The datasets have been collected from 220 residential buildings implemented in various cities in Egypt. Python was used to build the model during the training and testing phases. To evaluate the performance of the models generated by each algorithm, five key statistical metrics were employed. The findings revealed that both ANN and RF managed to predict the price per meter square of the construction projects, attaining model score accuracies of 96.6% and 93.05%, respectively. The best training for RF was acquired with 110 trees and a maximum decision tree depth of 20. The outcomes of this research could assist stakeholders in the building sector estimate construction project costs during the conceptual stage, enabling more informed decisions by examining key cost drivers for building projects and utilizing appropriate machine learning techniques for analysis. [ABSTRACT FROM AUTHOR]

In machine learning study proposes an ensemble-based strategy for both feature selection and data standardization to enhance model performance and interpretability. To maintain consistency across datasets, it employ average filling and weighted K-means clustering. Weighted K-means assigns distinct values to samples based on their distances to cluster centers, offering a more precise representation of the data distribution. Meanwhile, average filling replaces missing values with the average of corresponding features, ensuring a complete dataset for subsequent analysis. For feature selection, adopt an ensemble approach that combines Random Forest (RF) with Logistic Regression (LR) and ElasticNet. RF captures feature importance through tree-based analysis, while LR and ElasticNet provide additional insights into feature relevance and coefficients. This amalgamation aims to provide a comprehensive understanding of feature importance within the dataset. Principal Component Analysis (PCA) is employed to reduce dataset complexity while preserving key properties, facilitating more effective feature selection. By identifying orthogonal components that best explain data variation, PCA enables efficient representation and feature selection. In the final stage, Support Vector Machines (SVM) are utilized for categorization. SVM, a powerful classification method, establishes strong decision boundaries that optimize the gap between classes. Leveraging the selected features, the SVM model effectively categorizes new instances. [ABSTRACT FROM AUTHOR]

Using a random forest regression (RFR) machine learning technique, the critical temperature (T c) of a superconductor was predicted in the context of Industry 4.0 in this study using features derived from the material's physico-chemical properties, containing atomic mass, electron affinity, atomic radius, valence, and thermal conductivity. The same experimental data were also fitted with multilayer perceptron (MLP) artificial neural networks (ANN), M5 model tree and multivariate linear regression (MLR) model for comparison. The current investigation's findings show that the proposed RFR–relied model can successfully forecast the critical temperature of a superconductor. Additionally, the T c estimate was reached with a correlation coefficient of 0.9565 and a coefficient of determination 0.9146, when the observed dataset was used to test this unique technique. Additionally, the outcomes from the MLP, M5, and MLR models are obviously worse than those from the RFR–relied model. When it comes to fully comprehending the superconductivity, this investigation is noteworthy. Regarding forecasting effectiveness and feature reduction rate, the RFR approach has obvious advantages and generalizability, and it also demonstrates suitability for high-temperature superconductor T c forecasting. In fact, it offers a practical and affordable approach to data-driven superconductor investigation. [Display omitted] • The superconductor resistance is zero below its critical temperature (T c). • The developed model GS/RFR predicted satisfactorily the T c. • The GS/RFR model was compared with MLP and M5 tree techniques. • The correlation coefficient of the GS/RFR − relied model is about 0.92. • The relevance ranking of the input physico-chemical features is determined. [ABSTRACT FROM AUTHOR]

The study aims to compare the performance of various machine learning models for student persistence prediction. The research starts with a historical review of student retention studies and the evolution of predictive models in the field. It highlights the importance of predicting student persistence for educational institutions and individuals. It then describes a dataset from ResearchGate, consisting of anonymized undergraduate student data collected between 2008 and 2018, with 37 features and 4,424 records. Ten machine learning algorithms are considered, with two popular machine learning algorithms, Logistic Regression, and Random Forest classification, being compared in more detail for their performance in predicting student persistence. Evaluation metrics such as prediction accuracy, precision, recall, and F1- score are used. Results show that the Random Forest model outperforms Logistic Regression in predicting student outcomes, particularly when using the synthetic minority oversampling technique (SMOTE) to address the class imbalance. Overall, this study contributes to student retention research and provides insights for developing targeted support measures to enhance student success in higher education. [ABSTRACT FROM AUTHOR]

This paper presents a comprehensive methodology for precisely identifying smart home devices through the intricate analysis of network traffic data. The research framework comprises a multifaceted approach, encompassing critical stages such as in-depth exploratory data analysis, sophisticated feature engineering, robust machine learning modeling, iterative model refinement, and meticulous interpretation of results. The study evaluates various machine learning algorithms, with the Random Forest Classifier emerging as the optimal choice due to its superior performance in device recognition. Subsequently, the Random Forest Classifier's effectiveness is substantially bolstered through an exhaustive hyperparameter tuning procedure. This optimization effort culminates in a final accuracy score of 87.79%, demonstrating the excellent efficacy of this approach in precisely identifying smart home devices based on intricate network traffic patterns. Future research endeavors may concentrate on further refining these models and exploring additional feature sets to comprehensively augment their capabilities, thereby addressing the constantly evolving landscape of smart home technology and its associated security concerns. [ABSTRACT FROM AUTHOR]

Nosocomial infections represent an ongoing challenge to healthcare quality and patient safety, negatively impacting clinical outcomes and increasing the burden on healthcare systems. Thus, controlling this type of infection plays a very important role in ensuring a better quality of life for patients. Although the control and prevention measures for these infections are well defined, their signaling and detection is carried out manually and sometimes late, which compromises the health status of patients and everyone around them. In this context, this study emerged with the aim of exploring the potential of data mining techniques to predict the occurrence of nosocomial infections, with a specific focus on infections associated with surgical interventions. Using datasets for the period between 2018 and 2022, sourced from a Portuguese hospital and duly anonymized to protect patient privacy, several classification algorithms and data balancing techniques were analyzed to deal with the uneven nature of the data and the presence of minority classes. Among the algorithms and balancing techniques used, it was found that the Random Forest algorithm combined with the Oversampling technique showed superior performance in identifying cases of nosocomial infections associated with surgical interventions. The results of this study highlight the importance of collaboration between medicine and technology, indicating that the integration of data mining techniques can prove to be valuable tools to improve clinical decision-making and infection management in surgical context. [ABSTRACT FROM AUTHOR]

The article is devoted to the description of the real estate market, the situation in the market and services in the real estate market of Almaty and Kazakhstan as a whole. The factors influencing the market, such as external and internal, are considered, and further prospects for the development of the real estate market in the context of digitalization are presented The housing market in Kazakhstan is intricate, comprising numerous unique features among hundreds of thousands of apartments. It's important to note that even minor shifts in this market can trigger speculative behavior and intentional inflation of property prices. Thus, grasping the actual worth of an apartment and recognizing its prime location becomes crucial. The primary focus of the research was to construct a predictive model for real estate valuation using machine learning techniques. Various machine learning algorithms were employed, encompassing linear regression, lasso, ridge, decision tree regression, random forest regression, and SVM (utilizing a Gaussian kernel). The study illustrates the outcomes and precision of these algorithms, offering insight into their effectiveness. Based on the findings of this study, the Random Forest regression algorithm demonstrated the highest performance, achieving a 90% score. [ABSTRACT FROM AUTHOR]

In the rapidly evolving industrial landscape, Artificial Intelligence (AI) has emerged as a transformative force in various sectors, including manufacturing and supply chain management. Meanwhile, packaging planning is another area that is still open to AI development. Effective packaging planning is a complicated task to be handled carefully throughout the entire planning process. To solve this problem, we propose an improved heterogeneous chain-based multi-output classification model for predicting the dimensions and types of packages for each shipment. While conventional chain regression models typically employ only a single base classifier within each chain, our improved model allows for flexibility in the utilization of different classifiers within a chain structure. Our improved model is analyzed on a real-world dataset by employing different multi-output classification algorithms, including Random Forest (RF), Decision Trees (DT), and K-Nearest Neighbors (KNN). Experimental results demonstrate that our model, based solely on DT, outperforms by obtaining the best accuracy with a 0.98 overall accuracy value compared to traditional multi-output classification and chain regression models. Also, the proposed model with different classifiers has the second-best overall accuracy result in all models and has a higher overall accuracy result than traditional chain-based models. [ABSTRACT FROM AUTHOR]

Estimating an individual's height from a two-dimensional (2D) image has emerged as a focal point of investigation within Computer Vision. Notably, the research landscape has witnessed endeavors dedicated to human height estimation from frontal-face and full-body images. Nevertheless, there exists an imperative need for further exploration within this domain, primarily aimed at enhancing the precision of such estimations. Within this context, our study seeks to delve into the nuanced task of height from a solitary full-body image. Leveraging cutting-edge AI methodologies, we have harnessed the capabilities of Yolov7 for human pose estimation, alongside the DeepFace pre-trained model for age, gender, and race estimation, to serve as pivotal feature extraction mechanisms. Notably, our model's efficacy is intrinsically linked to the performance benchmarks set by these two foundational models. Our empirical observations underscore the commendable performance of Yolov7; however, it is incumbent to acknowledge that the DeepFace model has, regrettably, not demonstrated commensurate levels of accuracy in our experimentation. In the present study, we have employed a set of regression and tree-based predictive models, including Linear Regression, AdaBoost Regressor, Decision Tree, and Random Forest. Notably, the AdaBoost Regressor model has exhibited superior performance compared to the other models under consideration. Specifically, it has demonstrated a remarkable mean absolute error of 6.2 cms and a mean squared error of 7.9 cms, thereby establishing its efficacy in minimizing prediction errors in this context. [ABSTRACT FROM AUTHOR]

Titanium matrix composites (TMCs) offer superior specific mechanical properties compared to monolithic alloys. However, the complex interdependent effects of composition and processing on the resulting microstructure and properties make experimental determination of optimal TMC formulations challenging. This work explored a materials informatics approach integrating machine learning (ML) modeling with targeted fabrication and characterization for accelerated data-driven design of TMCs. A dataset of 368 data points on composition, processing method and mechanical properties of various TMCs was compiled from literature. Five ML regression algorithms were implemented to predict density, hardness and strength from composition-processing features. Among the models, random forest achieved highest accuracy with R2 scores above 0.93 and low errors. Fabrication of Ti-6Al-4 V/SiC using ML-guided parameters showed excellent agreement between predicted and experimentally measured properties. The ML models outperformed conventional empirical predictions by learning complex structure-property linkages from data. This integrated computational-experimental framework can guide rapid identification of property-optimized TMC formulations by reducing trial-and-error. Further work should focus on physics-based feature engineering and active learning. The data-driven approach demonstrated here shows promise for accelerating development of high-performance TMCs. [ABSTRACT FROM AUTHOR]

Detection and management of diabetes at an early stage is essential since it is rapidly becoming a global health crisis in many countries. Predictions of diabetes using machine learning algorithms have been promising. In this work, we use data collected from the Pima Indians to assess the performance of multiple machine-learning approaches to diabetes prediction. Ages, body mass indexes, and glucose levels for 768 patients are included in the data set. The methods evaluated are Logistic Regression, Decision Tree, Random Forest, k-Nearest Neighbors, Naive Bayes, Support Vector Machine, Gradient Boosting, and Neural Network. The findings indicate that the Logistic Regression and Neural Network models perform the best on most criteria when considering all classes together. The SVM, Random Forest, and Naive Bayes models also receive moderate to high scores, suggesting their strength as classification models. However, the kNN and Tree models show poorer scores on most criteria across all classes, making them less favorable choices for this dataset. The SGD, AdaBoost, and CN2 rule inducer models perform the poorest when comparing all models using a weighted average of class scores. The results of the study suggest that machine learning algorithms may help predict the onset of diabetes and for detecting the disease at an early stage. [ABSTRACT FROM AUTHOR]

The major Objective of this research paper is to design the Mobile Smart Device Digi Cash Intrusion Detection Framework (MSDDID) for assessing Intrusion Detection (ID) techniques and evaluating ID parameters that has to be rectified for enhancing the security of Digital Cash Transactions in Mobile Smart devices. The Research examined the Intrusion Detection dataset with 41 predictive features and 1 class feature for evaluating prediction in its novel form. The Framework was examined in WEKA with RapidMiner for analysis. The Results of classifiers Decision Table (98.7%), Random Forest Tree (99.79%), AdaBoost (94.37%), CART Model (99.61%), LazyIBK (99.44%), Naïve Bayesian (89.66%) signified that Smart devices security in Digi cash transactions could be predicted with refinement of data during transaction as deployed in this research work. The cluster analysis again conformed that num_root, su_attempted and num_compromised were the three parameters predominantly used for intrusions in the network and has to be addressed in the model. [ABSTRACT FROM AUTHOR]

Mental well-being is critical to an individual's health and quality of life. It encompasses emotional, psychological, and social dimensions, making it a complex and multifaceted construct. Traditionally, assessing and forecasting mental well-being has relied on self-report measures, clinical evaluations, and surveys, which can be subjective, time-consuming, and limited scope. To overcome these drawbacks and offer more precise and timely insights into mental well-being, this study presents a novel approach that uses the power of machine learning. This has been achieved by creating a comprehensive dataset containing various variables relevant to mental health. These variables include behavioral traits like exercise routines, sleep patterns, and social interactions and psychological traits like mood, stress levels, and emotional states. Environmental factors, such as geographic location, climate, and accessibility to mental health care, are also considered. A comprehensive understanding of the elements affecting mental well-being is made possible by this diversified dataset. The data is analyzed using machine learning models, such as deep learning neural networks, support vector machines, and random forests. These models were selected since they can capture intricate correlations and patterns in data, making them suitable for forecasting mental health. [ABSTRACT FROM AUTHOR]

BACKGROUND: Timely ventilator liberation can prevent morbidities associated with invasive mechanical ventilation in the pediatric ICU (PICU). There currently exists no standard benchmark for duration of invasive mechanical ventilation in the PICU. This study sought to develop and validate a multi-center prediction model of invasive mechanical ventilation duration to determine a standardized duration of invasive mechanical ventilation ratio. METHODS: This was a retrospective cohort study using registry data from 157 institutions in the Virtual Pediatric Systems database. The study population included encounters in the PICU between 2012-2021 involving endotracheal intubation and invasive mechanical ventilation in the first day of PICU admission who received invasive mechanical ventilation for > 24 h. Subjects were stratified into a training cohort (2012-2017) and 2 validation cohorts (2018-2019/2020-2021). Four models to predict the duration of invasive mechanical ventilation were trained using data from the first 24 h, validated, and compared. RESULTS: The study included 112,353 unique encounters. All models had observed-to-expected (O/E) ratios close to one but low mean squared error and R² values. The random forest model was the best performing model and achieved an O/E ratio of 1.043 (95% CI 1.030-1.056) and 1.004 (95% CI 0.990-1.019) in the validation cohorts and 1.009 (95% CI 1.004-1.016) in the full cohort. There was a high degree of institutional variation, with single- unit O/E ratios ranging between 0.49-1.91. When stratified by time period, there were observ- able changes in O/E ratios at the individual PICU level over time. CONCLUSIONS: We derived and validated a model to predict the duration of invasive mechanical ventilation that performed well in aggregated predictions at the PICU and the cohort level. This model could be beneficial in quality improvement and institutional benchmarking initiatives for use at the PICU level and for tracking of performance over time. [ABSTRACT FROM AUTHOR]

Smart Grid is an intelligent power grid with a bidirectional flow of electricity and information, that applies intelligent techniques to operate the grid autonomously near the stability limit. An intelligent technique is developed to identify and predict the abnormalities due to changes in customer behaviour and the unexpected disruption in the grid. A cost-sensitive stacked ensemble classifier (CS-SEC) is proposed for predicting the operations in smart grid that combines four cost-sensitive base classifiers, namely Extreme gradient boosting, Naive Bayes, Nu-support vector machine and Random forest at level-1 and the support vector machine as the meta classifier in level-2. The meta classifier uses the probability of prediction of the first-level classifiers with stratified 5-fold cross-validation to predict the decentralized smart grid stability. The proposed stacked ensemble classifier achieved an accuracy of 98.6% with specificity, recall and precision of 98.34%, 99.0% and 99.06%, respectively. Extensive experimental evaluation and results show that the proposed CS-SEC provides an accurate prediction of grid stability compared with other state-of-the-art models. The results reveal the robustness and competency of the proposed CS-SECs with optimized parameters. [ABSTRACT FROM AUTHOR]

The main purpose of this study was to compare the performance and validation of six machine learning models (extreme gradient boosting, random forest, artificial neural network, support vector machine, C4.5 decision tree, and naive Bayes) in landslide susceptibility modelling. The province of Rize, which has the highest rate of landslide events in Türkiye, was chosen as the study area. The conditioning factors (distance to roads, lithology, drainage density, slope, topographic wetness index (TWI), soil depth, distance to rivers, land use, NDVI, plan curvature, elevation, aspect, profile curvature) affecting the landslide were determined using the ReliefF method. A total of 516 landslides were identified for creating models, comparing performance, and validating results. The performance and validation of the models were determined by the receiver operating characteristics (ROC), sensitivity, specificity, accuracy, and kappa index. The results show that the XGBoost model outperforms the other five machine learning models in terms of accuracy and performance and is the most effective model for generating landslide susceptibility maps in Rize (Türkiye). [ABSTRACT FROM AUTHOR]

Stroke is the leading cause of permanent disability in adults, and it can cause permanent brain damage. According to the World Health Organization, 795 000 Americans experience a new or recurrent stroke each year. Early detection of medical disorders, for example, strokes, can minimize the disabling effects. Thus, in this paper, we consider various risk factors that contribute to the occurrence of stoke and machine learning algorithms, for example, the decision tree, random forest, and naive Bayes algorithms, on patient characteristics survey data to achieve high prediction accuracy. We also consider the semisupervised self‐training technique to predict the risk of stroke. We then consider the near‐miss undersampling technique, which can select only instances in larger classes with the smaller class instances. Experimental results demonstrate that the proposed method obtains an accuracy of approximately 98.83% at low cost, which is significantly higher and more reliable compared with the compared techniques. [ABSTRACT FROM AUTHOR]

Modulation classification is a crucial technique to utilize the unconsumed spectrum in Cognitive Radio (CR) and Dynamic Spectrum Access (DSA) systems to meet the required traffic demands for future-generation cellular networks. This paper presents an end-to-end experimental setup as a generic methodology to implement various Transfer Learning (TL) models in an indoor environment. This allows us to learn the features from multiple modulation signals to train and test the model. The performance evaluation of proposed TL models such as Convolutional Neural Network - Random Forest (CNN-RF), and Convolutional Long Short Term Deep Neural Network (CLDNN) - Random Forest (CLDNN-RF) have been thoroughly discussed. The result shows that the proposed TL models yield more than 90% classification accuracy for various modulation types. A proposed framework for location-specific TL model selection based on the maximum classification accuracy has been investigated. [ABSTRACT FROM AUTHOR]

The drug development process takes a long time since it requires sorting through a large number of inactive compounds from a large collection of compounds chosen for study and choosing just the most pertinent compounds that can bind to a disease protein. The use of virtual screening in pharmaceutical research is growing in popularity. During the early phases of medication research and development, it is crucial. Chemical compound searches are nowmore narrowly targeted. Because the databases containmore andmore ligands, thismethod needs to be quick and exact. Neural network fingerprints were created more effectively than the well-known Extended Connectivity Fingerprint (ECFP). Only the largest sub-graph is taken into consideration to learn the representation, despite the fact that the conventional graph network generates a better-encoded fingerprint. When using the average or maximum pooling layer, it also contains unrelated data. This article suggested the Graph Convolutional Attention Network (GCAN), a graph neural network with an attention mechanism, to address these problems. Additionally, it makes the nodes or sub-graphs that are used to create the molecular fingerprint more significant. The generated fingerprint is used to classify drugs using ensemble learning. As base classifiers, ensemble stacking is applied to Support Vector Machines (SVM), Random Forest, Nave Bayes, Decision Trees, AdaBoost, and Gradient Boosting. When compared to existing models, the proposed GCAN fingerprint with an ensemble model achieves relatively high accuracy, sensitivity, specificity, and area under the curve. Additionally, it is revealed that our ensemble learning with generated molecular fingerprint yields 91% accuracy, outperforming earlier approaches. [ABSTRACT FROM AUTHOR]

Measuring and classifying the water quality is necessary to manage the appropriate water quality for various farms near the coast or affected by seawater. This research aimed to improve the water quality classification model for various farms using Multi-Layer Perceptron Neural Network-based multiclass Support Vector Machine. It also implements the Random Forest Feature Importance Selection to increase model accuracy. The class reduction technique decreases the probability of co-occurrence classes for various farms in overlapping water ecosystems. The result has shown that the dataset that applied the class reduction helped increase the model's efficiency more than the feature selection technique. The models that applied the multi-class Support Vector Machine classifier are more accurate than the Softmax activation function classifier. The findings indicate that the model using Multi-Layer Perceptron Neural Network-based One-versus-One Support Vector Machine combined with the Random Forest Feature Importance Selection and the class reduction has the highest efficiency and improves the water quality classification model in various farms. [ABSTRACT FROM AUTHOR]

Technology provides various conveniences for users in many aspects, such as in the commercial business world with the development of financial technology or fintech. In Indonesia, there are fintech services with the most users, namely OVO and DANA. Various kinds of criticisms and suggestions related to deficiencies and weaknesses in OVO and DANA services were submitted by users via social media Twitter. This study aims to analyze and develop a sentiment identification model for user complaints against OVO and DANA services on Twitter. This study uses the four classification algorithm methods: Random Forest, K-NN, Decision Tree, and Naïve Bayes Classifier. The data from social media Twitter is taken from user tweets regarding criticism and complaints on OVO and DANA services. The data obtained was 5,000 instances consisting of 20% positive and 80% negative reviews. Accuracy values represent the study results with 87% for Naive Bayes, 86% for Decision Tree, 91% for K-NN, and 87% for Random Forest. Based on these results, it can be concluded that testing of sentiment analysis on user complaints on OVO and DANA services using the K-NN algorithm is superior to the other algorithms. This research results in a system that combines the four prediction models. The evaluation shows that the developed model can accurately classify user sentiment. [ABSTRACT FROM AUTHOR]

• A systematic evaluation of four machine learning algorithms were explored to rationalise the complex relationships in Mg-alloy data. • The random forest regression model and neural network were robust models for predicting Mg alloys properties. • A graphical user interface (GUI) webtool was developed to facilitate the use of the machine learning models in predicting the mechanical properties of Mg alloys. • Machine learning (ML) models provide significant implications for the design and development of high-performance Mg alloys. Machine learning (ML) models provide great opportunities to accelerate novel material development, offering a virtual alternative to laborious and resource-intensive empirical methods. In this work, the second of a two-part study, an ML approach is presented that offers accelerated digital design of Mg alloys. A systematic evaluation of four ML regression algorithms was explored to rationalise the complex relationships in Mg-alloy data and to capture the composition-processing-property patterns. Cross-validation and hold-out set validation techniques were utilised for unbiased estimation of model performance. Using atomic and thermodynamic properties of the alloys, feature augmentation was examined to define the most descriptive representation spaces for the alloy data. Additionally, a graphical user interface (GUI) webtool was developed to facilitate the use of the proposed models in predicting the mechanical properties of new Mg alloys. The results demonstrate that random forest regression model and neural network are robust models for predicting the ultimate tensile strength and ductility of Mg alloys, with accuracies of ∼80% and 70% respectively. The developed models in this work are a step towards high-throughput screening of novel candidates for target mechanical properties and provide ML-guided alloy design. [ABSTRACT FROM AUTHOR]

Nowadays, wood identification is made by experts using hand lenses, wood atlases, and field manuals which take a lot of cost and time for the training process. The quantity and species must be strictly set up, and accurate identification of the wood species must be made during exploitation to monitor trade and enforce regulations to stop illegal logging. With the development of science, wood identification should be supported with technology to enhance the perception of fairness of trade. An automatic wood identification system and a dataset of 50 commercial wood species from Asia are established, namely, wood anatomical images collected and used to train for the proposed model. In the convolutional neural network (CNN), the last layers are usually soft-max functions with dense layers. These layers contain the most parameters that affect the speed model. To reduce the number of parameters in the last layers of the CNN model and enhance the accuracy, the structure of the model should be optimized and developed. Therefore, a hybrid of convolutional neural network and random forest model (CNN-RF model) is introduced towood identification. The accuracy's hybrid model ismore than 98%, and the processing speed is 3 times higher than the CNN model. The highest accuracy is 1.00 in some species, and the lowest is 0.92. These results show the excellent adaptability of the hybrid model in wood identification based on anatomical images. It also facilitates further investigations of wood cells and has implications for wood science. [ABSTRACT FROM AUTHOR]

During the operation of photovoltaic systems, various faults can occur and result in serious problems, such as energy loss or system shutdown. Therefore, it is crucial to identify and diagnose these faults in order to improve system performance. The purpose of this work is to propose an efficient and simple procedure for the early detection and diagnosis of faults on the direct current side of photovoltaic systems. These faults include the short circuit of three modules, short circuit of ten modules, and string disconnection. Therefore, it is necessary to distinguish between four classes: the healthy class and three classes representing different types of faults. A dataset representing the four classes and comprising four measured attributes—cell temperature, solar irradiance, and current and voltage at the maximum power point—is utilized in the developed approach. The idea is to transform the multiclassification problem into a binary classification problem and utilize a modified version of the well-known K-nearest neighbors (KNN) classifier. In this proposed version, the training dataset is divided into two hyperspheres, each representing a distinct class. The Giza pyramid construction algorithm is then utilized to determine the optimal center coordinates of these hyperspheres. To classify a new data point using the proposed classifier, which combines the KNN classifier and the Giza pyramid construction algorithm, distances are computed only between the new data point and the center of each sphere. Unlike the classical version of the KNN classifier, which involves computing distances between the new data point and the entire dataset. To assess the efficiency of the proposed approach, a comparative study was conducted, including the classical version of the KNN, support vector machine, decision tree, and random forest algorithms. The evaluation criteria considered were accuracy, precision, recall, and execution time. The results of the carried-out study demonstrated the remarkable superiority of the proposed algorithm over these alternative methods. [ABSTRACT FROM AUTHOR]

The rapid growth of Fintech innovation and the widespread adoption of blockchain technologies have indeed had a transformative impact on the financial industry. In this paper, the focus is on the application of machine learning algorithms, specifically the Random Forest Regression algorithm, within the context of Fintech and blockchain. This research contributes to the advancement of machine learning techniques in the field of Fintech and blockchain. The Random Forest Regression algorithm utilizes ensemble learning, combining multiple decision trees to analyze complex financial data and make predictions on various outcomes. This algorithm has proven to be effective in addressing key challenges within the industry, such as predicting loan defaults, detecting fraud, and assessing risks. Through experimental evaluations and case studies, the paper demonstrates the effectiveness of the Random Forest Regression algorithm in enhancing Fintech innovation in blockchain applications. The algorithm improved accuracy, scalability, and interpretability enable financial institutions to make data-driven decisions and optimize their operations. [ABSTRACT FROM AUTHOR]

Credit card Fraud detection is a critical task in various industries, including finance and e-commerce, where identifying fraudulent activities can help prevent financial losses and protect users. It begins by combining two datasets containing fraudulent and non-fraudulent transactions to create a comprehensive dataset for analysis. Data is preprocessed by removing unnecessary features, calculating distance metrics, and generating new variables to capture temporal patterns and transaction history. Multicollinearity issues are addressed through feature selection. Improved Random Forest (RF) algorithm is used to improve fraud detection. The experimental results indicate that the improved Random Forest algorithm achieves commendable accuracy in fraud detection. The proposed model achieves 99.87% training accuracy and 99.41% testing accuracy. The Model's performance is evaluated by measuring precision, recall, F1-score and support. Our research emphasizes the importance of considering improved algorithms to achieve better results. The findings provide valuable insights for organizations aiming to enhance their fraud detection capabilities and make informed decisions to protect their systems and users. [ABSTRACT FROM AUTHOR]

Objective: Streamflow is a primary determinant of fish assemblage structure in riverine systems, but alteration of natural flow regimes can result in fish assemblage shifts through the process of environmental filtering. Because natural drought conditions reduce and homogenize streamflow in a manner comparable to projections for climate change in some regions, drought could serve as a proxy for expected future flow conditions. Methods: We investigated the effects of drought as a temporally dynamic environmental filter of the occurrence of two guilds of fishes, benthic spawners that deposit adhesive ova along the benthic zone of rivers and pelagic spawners that release semibuoyant, nonadhesive ova into the pelagic zone of rivers. We developed species‐specific random forest models to estimate annual probability of occurrence for three benthic‐spawning and four pelagic‐spawning minnow species at three sites in the upper Brazos River, Texas, for the period 1950–2018. We then used a generalized additive mixed‐effects model to assess the relationship between drought intensity and likelihood of occurrence to test whether reproductive modes differed in response to drought (hypothesis 1) and whether response to drought was spatially (hypothesis 2) or temporally (hypothesis 3) variable. Result: We found support for hypothesis 1 as two of four pelagic‐spawning species (Shoal Chub Macrhybopsis hyostoma, Smalleye Shiner Notropis buccula) declined and two of three benthic‐spawning species increased as drought intensified, support for hypothesis 2 as responses to drought varied by gauge location, and no support for hypothesis 3 as drought response was consistent for periods 1950–1979 and 1980–2018. Conclusion: These findings offer insight into the future of riverine fish assemblages as climate change is expected to exacerbate regional drought conditions. Management of pelagic‐spawning fishes during extreme drought in the southern Great Plains may require strategies such as (1) rescues of fish from drying reaches, (2) captive holding and propagation, and (3) assisted recolonization following subsidence of drought conditions. Impact statementExtreme drought caused a temporal increase in occurrence of benthic‐spawning fishes and decrease in occurrence of pelagic‐spawning fishes. Management to conserve imperiled pelagic‐spawning fishes should consider the effects of drought even within otherwise suitable habitats. [ABSTRACT FROM AUTHOR]

Credit risk has become an issue for banks and financial institutions as default payments from customers result in lost funds for these companies. Machine learning techniques are being used to assist financial institutions in determining clients with the highest probability of repaying their loans. This study determined factors associated with high-risk clients based on demographic data and credit history. Three supervised learning algorithms were explored in classifying clients into high and low risk categories including logistic regression, random forest, and k-nearest neighbor. Models were evaluated for performance using the confusion matrix technique. Results showed that Random Forest was the best fit model for this study. [ABSTRACT FROM AUTHOR]

In today's increasingly competitive job market, many people have turned to entrepreneurship as a second career option. How to determine whether one's entrepreneurship can be successful has become a major topic of discussion. However, there is currently limited research on identifying the success and failure of entrepreneurship. Therefore, this paper used a questionnaire to obtain the characteristics of successful entrepreneurs, selected different machine learning algorithms to construct models for identifying the success or failure of entrepreneurs, and analyzed these models with a dataset. It was found that, under different evaluation criteria, the random forest (RF) model had the best accuracy and kappa value, 0.8300 and 0.6520, respectively, for the overall identification effect of the model. From the perspective of identifying the successful entrepreneur group, the precision and recall rate of the K-nearest neighbor model were the highest, 0.9215 and 0.8630, respectively. From the perspective of identifying the failed entrepreneurial group, the specificity of the RF model was the lowest, 0.8054. Therefore, under the comprehensive consideration of evaluation indicators, the RF model had the best overall identification effect among the three identification models. [ABSTRACT FROM AUTHOR]

To protect the environment, the discharged sewage's quality must meet the state's discharge standards. There are many water quality indicators, and the pH (Potential of Hydrogen) value is one of them. The natural water's pH value is 6.0-8.5. The sewage treatment plant uses some data in the sewage treatment process to monitor and predict whether wastewater's pH value will exceed the standard. This paper aims to study the deep learning prediction model of wastewater's pH. Firstly, the research uses the random forest method to select the data features and then, based on the sliding window, convert the data set into a time series which is the input of the deep learning training model. Secondly, by analyzing and comparing relevant references, this paper believes that the CNN (Convolutional Neural Network) model is better at nonlinear data modeling and constructs a CNN model including the convolution and pooling layers. After alternating the combination of the convolutional layer and pooling layer, all features are integrated into a full-connected neural network. Thirdly, the number of input samples of the CNN model directly affects the prediction effect of the model. Therefore, this paper adopts the sliding window method to study the optimal size.Many experimental results show that the optimal prediction model can be obtained when alternating six convolutional layers and three pooling layers. The last full-connection layer contains two layers and 64 neurons per layer. The sliding window size selects as 12. Finally, the research has carried out data prediction based on the optimal CNN deep learning model. The predicted pH of the sewage is between 7.2 and 8.6 in this paper. The result is applied in the monitoring system platform of the "Intelligent operation and maintenance platform of the reclaimed water plant". [ABSTRACT FROM AUTHOR]

Cardiotocography (CTG) represents the fetus's health inside the womb during labor. However, assessment of its readings can be a highly subjective process depending on the expertise of the obstetrician. Digital signals from fetal monitors acquire parameters (i.e., fetal heart rate, contractions, acceleration). Objective: This paper aims to classify the CTG readings containing imbalanced healthy, suspected, and pathological fetus readings. Method: We perform two sets of experiments. Firstly, we employ five classifiers: Random Forest (RF), Adaptive Boosting (AdaBoost), Categorical Boosting (CatBoost), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LGBM) without over-sampling to classify CTG readings into three categories: healthy, suspected, and pathological. Secondly, we employ an ensemble of the above-described classifiers with the oversamplingmethod. We use a random over-sampling technique to balance CTG records to train the ensemble models. We use 3602 CTG readings to train the ensemble classifiers and 1201 records to evaluate them. The outcomes of these classifiers are then fed into the soft voting classifier to obtain the most accurate results. Results: Each classifier evaluates accuracy, Precision, Recall, F1-scores, and Area Under the Receiver Operating Curve (AUROC) values. Results reveal that the XGBoost, LGBM, and CatBoost classifiers yielded 99% accuracy. Conclusion: Using ensemble classifiers over a balanced CTG dataset improves the detection accuracy compared to the previous studies and our first experiment. A soft voting classifier then eliminates the weakness of one individual classifier to yield superior performance of the overall model. [ABSTRACT FROM AUTHOR]

COVID-19 has significantly impacted the growth prediction of a pandemic, and it is critical in determining how to battle and track the disease progression. In this case, COVID-19 data is a time-series dataset that can be projected using different methodologies. Thus, this work aims to gauge the spread of the outbreak severity over time. Furthermore, data analytics and Machine Learning (ML) techniques are employed to gain a broader understanding of virus infections. We have simulated, adjusted, and fitted several statistical time-series forecasting models, linearML models, and nonlinear ML models. Examples of these models are Logistic Regression, Lasso, Ridge, ElasticNet, Huber Regressor, Lasso Lars, Passive Aggressive Regressor, K-Neighbors Regressor, Decision Tree Regressor, Extra Trees Regressor, Support Vector Regressions (SVR), AdaBoost Regressor, Random Forest Regressor, Bagging Regressor, AuoRegression, MovingAverage, Gradient Boosting Regressor, Autoregressive Moving Average (ARMA), Auto-Regressive Integrated Moving Averages (ARIMA), SimpleExpSmoothing, Exponential Smoothing, Holt-Winters, Simple Moving Average, Weighted Moving Average, Croston, and naive Bayes. Furthermore, our suggested methodology includes the development and evaluation of ensemble models built on top of the best-performing statistical and ML-based prediction methods. A third stage in the proposed system is to examine three different implementations to determine which model delivers the best performance. Then, this best method is used for future forecasts, and consequently, we can collect the most accurate and dependable predictions. [ABSTRACT FROM AUTHOR]