Analysis Of Machine Learning Techniques Research Articles

Comparative analysis of machine learning techniques in predicting dielectric behavior of ternary chalcogenide thin films

Abstract The current research introduces a comparative analysis of the dielectric behavior exhibited by ternary chalcogenide thin films, including both experimental data and machine learning techniques. The study of temperature and frequency dependencies of dielectric parameters is crucial for assessing material losses, particularly focusing on the low-frequency range where dielectric dispersion occurs. The experimental results on the frequency (100–1000000 Hz)and the temperature(303–393 K) influences on the dielectric (constant ( ε 1 ) and loss ( ε 2 )) of A s 4 G e 24 T e 72 composition in thin film form have been discussed. Results demonstrate that ε 1 exhibits an increasing trend with temperature, attributed to heightened orientation polarization as dipoles gain mobility with elevated thermal energy. Conversely, ε 1 declines with rising frequency, reflecting diverse different polarization mechanisms. Understanding these behaviors is important for material characterization and applications in fields like electronics and solar cells. A comparative analysis of the dielectric behavior exhibited by ternary chalcogenide thin films, including both experimental data and machine learning techniques. Through the experimental approach, the dielectric properties of the films are investigated. Additionally, the study employs a range of machine learning algorithms, including Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference Systems (ANFIS), Genetic Programming (GP), hybrid techniques ANFIS-GA, and ANFIS-PSO, to model and predict the dielectric behavior with remarkable accuracy. The paper presents a comparison between the performance of the proposed machine learning models, highlighting their strengths and limitations in capturing the complex dielectric phenomena observed in the ternary chalcogenide thin films. This comparative study not only provides valuable insights into the dielectric properties of these materials but also demonstrates the efficacy of machine learning in understanding and predicting their behavior, paving the way for further advancements in materials science and device development.

A Multivariate Phenotypical Approach of Sepsis and Septic Shock—A Comprehensive Narrative Literature Review

Despite medical advances, sepsis and septic shock remain some of the leading causes of mortality worldwide, with a high inter-individual variability in prognosis, clinical manifestations and response to treatment. Evidence suggests that pulmonary sepsis is one of the most severe forms of sepsis, while liver dysfunction, left ventricular dysfunction, and coagulopathy impact the prognostic. Sepsis-related hypothermia and a hypoinflammatory state are related to a poor outcome. Given the heterogeneity of sepsis and recent technological progress amongst machine learning analysis techniques, a new, personalized approach to sepsis is being intensively studied. Despite the difficulties when tailoring a targeted approach, with the use of artificial intelligence-based pattern recognition, more and more publications are becoming available, highlighting novel factors that may intervene in the high heterogenicity of sepsis. This has led to the devise of a phenotypical approach in sepsis, further dividing patients based on host and trigger-related factors, clinical manifestations and progression towards organ deficiencies, dynamic prognosis algorithms, and patient trajectory in the Intensive Care Unit (ICU). Host and trigger-related factors refer to patients’ comorbidities, body mass index, age, temperature, immune response, type of bacteria and infection site. The progression to organ deficiencies refers to the individual particularities of sepsis-related multi-organ failure. Finally, the patient’s trajectory in the ICU points out the need for a better understanding of interindividual responses to various supportive therapies. This review aims to identify the main sources of variability in clustering septic patients in various clinical phenotypes as a useful clinical tool for a precision-based approach in sepsis and septic shock.

Comparative Analysis of Machine Learning Techniques for Identifying Multiple Force Systems from Accelerometer Measurements.

The knowledge of the forces acting on a structure enables, among many other factors, assessments of whether the component's useful life is compromised by the current machine condition. In many cases, a direct measurement of those forces becomes unfeasible, and an inverse problem must be solved. Among the solutions developed, machine learning techniques have stood out as powerful predictive tools increasingly applied to engineering problem-solving. This study evaluates the ability of different machine learning models to identify parameters of multi-force systems from accelerometer measurements. The models were assessed according to their prediction potential based on correlation coefficient (R2), mean relative error (MRE), and processing time. A computational numerical model using the finite element method was generated and validated by vibration measurements performed using accelerometers in the laboratory. A robust database created by the response surface methodology in conjunction with Design of Experiment (DOE) was used for the evaluation of the ability of machine learning models to predict the position, frequency, magnitude, and number of forces acting on a structure. Among the six machine learning models evaluated, k-NN was able to predict with a 0.013% error, and Random Forests showed a maximum error of 0.2%. The innovation of this study lies in the application of the proposed method for identifying parameters of multi-force systems.

Comparative Analysis of Machine Learning Techniques for Sentiment Classification

Abstract—Sentiment analysis is an important part of natural language processing (NLP) and has many applications in social media, e-commerce, and other fields. This study aims to provide a clear distinction between these methodologies by offering a structured overview of sentiment analysis and the variety of techniques used in its execution. The article examines different machine learning algorithms for sentiment analysis and high- lights their advantages and disadvantages by drawing on credible prior research on the subject. Additionally, the study provides a tabular comparison of different machine learning methods by selecting suitable parameters. Keywords: Decision Tree, Support Vector Machine (SVM), Random Forest, Convolution neural network (CNN), Neural network, Long Short-Term Memory Networks (LSTM), BERT

A Comparative Analysis of Machine Learning Techniques for Predicting the Performance of Microchannel Gas Coolers in CO2 Automotive Air-Conditioning Systems

The automotive industry is increasingly focused on developing more energy-efficient and eco-friendly air-conditioning systems. In this context, CO2 microchannel gas coolers (MCGCs) have emerged as promising alternatives due to their low global warming potential (GWP) and environmental benefits. This paper explores the application of machine learning (ML) algorithms to predict the thermohydraulic performance of MCGCs in automotive air-conditioning systems. Using data generated from an experimentally validated numerical model, this study compares various ML techniques, including both linear and nonlinear regression models, to forecast key performance metrics such as refrigerant outlet temperature, pressure drop, and heat transfer rate. Spearman’s correlation was employed to develop performance maps, whereas the R2 and MSE metrics were used to evaluate the models’ predictive accuracy. The linear models gave around 70% forecasting accuracy for pressure drop across the gas cooler and 97% accuracy for refrigerant outlet temperature, whereas the nonlinear models achieved more accurate predictions, with an accuracy ranging from 71% to 99%. This implies that nonlinear regression generally performs better than linear regression models in assessing the overall thermohydraulic performance of microchannel gas coolers. This research brings forth new ideas on how ML methods can be applied to enhance efficiency and effectiveness in gas coolers, contributing to the development of more eco-friendly automotive air-conditioning systems.

Prediction of Flotation Deinking Performance: A Comparative Analysis of Machine Learning Techniques

Flotation deinking is one of the most widely used techniques for the separation of ink particles from cellulose fibers during the process of paper recycling. It is a complex process influenced by a variety of factors, and is difficult to represent and usually results in models that are inconvenient to implement and/or interpret. In this paper, a comprehensive study of several machine learning methods for the prediction of flotation deinking performance is carried out, including support vector regression, regression tree ensembles (random forests and boosting) and Gaussian process regression. The prediction relies on the development of a limited dataset that assumes representative data samples obtained under a variety of laboratory conditions, including different reagents, pH values and flotation residence times. The results obtained in this paper confirm that the machine learning methods enable the accurate prediction of flotation deinking performance even when the dataset used for training the model is limited, thus enabling the determination of optimal conditions for the paper recycling process, with only minimal costs and effort. Considering the low complexity of the Gaussian process regression compared to the aforementioned ensemble models, it should be emphasized that the Gaussian process regression gave the best performance in estimating fiber recovery (R2 = 97.77%) and a reasonable performance in estimating the toner recovery (R2 = 86.31%).

Comparative analysis of machine learning techniques for detecting potability of water

This research aims to evaluate the effectiveness of machine learning algorithms in determining the potability of water. In the study, a total of 3276 water samples were analyzed for 10 different features that determine the potability of water. Besides that, the study's consideration is to evaluate the impact of trimming, IQR, and percentile methods on the performance of machine learning algorithms. The models were built using nine different classification algorithms (Logistic Regression, Decision Trees, Random Forest, XGBoost, Naive Bayes, K-Nearest Neighbors, Support Vector Machine, AdaBoost, and Bagging Classifier). According to the results, filling the missing data with the population mean and handling outliers with Trimming and IQR methods improved the performance of the models. Random Forest and Decision Tree algorithms were the most accurate in determining the potability of water. The findings of this research are of high importance to sustainable water resource management and serve as a crucial input for the decision-making process on the quality of water. The study also offers an example for researchers working on datasets that contain missing values and outliers.

Comparative analysis of machine learning techniques for enhanced vehicle tracking and analysis

The past few years have seen a marvellous growth in technology and science. This rapid improvement has proven to be a blessing, making human life easier. Technological developments such as autonomous driving systems and electric cars have made it easier to travel in a dependable and economical manner, satisfying the increasing need for convenient and environmentally friendly travel. However, the increase in traffic has led to a surge in accidents and road casualties. Despite efforts to enhance automobile design and traffic control, there remains a significant need for implementing a system for vehicle tracking, accident detection, and notification. Delays in information and unfulfilled medical needs often result in the loss of lives following accidents. This study reviews and compares different automatic accident detection and notification systems that use accelerometers, vibration detectors, and GPS technology to notify registered contacts of an accident's location via SMS or email. The analysis that follows will specifically look at the benefits, drawbacks, and future uses of various technologies that are used in these systems. In this study, different machine learning-based methods for improving the accuracy of car tracking and cutting down on reaction times in accident situations will be looked at and compared. For testing their usefulness, we used deep learning models like CNN, SVM, and YOLOv3 on a number of different datasets. According to our data, these methods greatly enhance the accuracy of spotting, with YOLOv3 showing the best level of accuracy. Furthermore, the study talks about the pros, cons, and possible future uses of these technologies. It stresses the need for more research into improving model performance in different situations.

Comparative Analysis of Machine Learning Techniques for Water Consumption Prediction: A Case Study from Kocaeli Province.

This study presents a comparative analysis of various Machine Learning (ML) techniques for predicting water consumption using a comprehensive dataset from Kocaeli Province, Turkey. Accurate prediction of water consumption is crucial for effective water resource management and planning, especially considering the significant impact of the COVID-19 pandemic on water usage patterns. A total of four ML models, Artificial Neural Networks (ANN), Random Forest (RF), Support Vector Machines (SVM), and Gradient Boosting Machines (GBM), were evaluated. Additionally, optimization techniques such as Particle Swarm Optimization (PSO) and the Second-Order Optimization (SOO) Levenberg-Marquardt (LM) algorithm were employed to enhance the performance of the ML models. These models incorporate historical data from previous months to enhance model accuracy and generalizability, allowing for robust predictions that account for both short-term fluctuations and long-term trends. The performance of each model was assessed using cross-validation. The R2 and correlation values obtained in this study for the best-performing models are highlighted in the results section. For instance, the GBM model achieved an R2 value of 0.881, indicating a strong capability in capturing the underlying patterns in the data. This study is one of the first to conduct a comprehensive analysis of water consumption prediction using machine learning algorithms on a large-scale dataset of 5000 subscribers, including the unique conditions imposed by the COVID-19 pandemic. The results highlight the strengths and limitations of each technique, providing insights into their applicability for water consumption prediction. This study aims to enhance the understanding of ML applications in water management and offers practical recommendations for future research and implementation.

Empirical evaluation of feature selection methods for machine learning based intrusion detection in IoT scenarios

This paper delves into the critical need for enhanced security measures within the Internet of Things (IoT) landscape due to inherent vulnerabilities in IoT devices, rendering them susceptible to various forms of cyber-attacks. The study emphasizes the importance of Intrusion Detection Systems (IDS) for continuous threat monitoring. The objective of this study was to conduct a comprehensive evaluation of feature selection (FS) methods using various machine learning (ML) techniques for classifying traffic flows within datasets containing intrusions in IoT environments. An extensive benchmark analysis of ML techniques and FS methods was performed, assessing feature selection under different approaches including Filter Feature Ranking (FFR), Filter-Feature Subset Selection (FSS), and Wrapper-based Feature Selection (WFS). FS becomes pivotal in handling vast IoT data by reducing irrelevant attributes, addressing the curse of dimensionality, enhancing model interpretability, and optimizing resources in devices with limited capacity. Key findings indicate the outperformance for traffic flows classification of certain tree-based algorithms, such as J48 or PART, against other machine learning techniques (naive Bayes, multi-layer perceptron, logistic, adaptive boosting or k-Nearest Neighbors), showcasing a good balance between performance and execution time. FS methods' advantages and drawbacks are discussed, highlighting the main differences in results obtained among different FS approaches. Filter-feature Subset Selection (FSS) approaches such as CFS could be more suitable than Filter Feature Ranking (FFR), which may select correlated attributes, or than Wrapper-based Feature Selection (WFS) methods, which may tailor attribute subsets for specific ML techniques and have lengthy execution times. In any case, reducing attributes via FS has allowed optimization of classification without compromising accuracy. In this study, F1 score classification results above 0.99, along with a reduction of over 60% in the number of attributes, have been achieved in most experiments conducted across four datasets, both in binary and multiclass modes. This work emphasizes the importance of a balanced attribute selection process, taking into account threat detection capabilities and computational complexity.

Comparative analysis of machine learning techniques for accurate prediction of unfrozen water content in frozen soils

Unfrozen water content (UWC) plays a critical role in determining the thermal, hydraulic, and mechanical properties of frozen soils. Existing empirical, semi-empirical, and theoretical models for UWC estimation have limitations in terms of accuracy as well as generalizability. To address these challenges, the present study explored the application of six machine learning techniques to predict UWC in frozen soils: Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), K-Nearest Neighbors (KNN), Support Vector Regression (SVR), and Backpropagation Neural Network (BPNN). Considering the UWC hysteresis phenomenon between the freezing and thawing processes, experimental UWC data collected from the literature were separated into two sub-datasets: freezing branch dataset (FBD) and thawing branch dataset (TBD). Based on that, a comprehensive framework integrating Bayesian optimization and 10-fold cross-validation was established to optimize the six models' hyperparameters and to evaluate their performance. The results highlighted significant variations in the predictive capability among the six machine learning models, with ensemble methods (i.e., RF, XGBoost, LightGBM) generally demonstrating superior accuracy. Feature importance analysis, robustness checks, and uncertainty quantification further elucidated the strengths and limitations of each model. The present study provides profound insights into the selection and application of machine learning models for accurately modeling the properties of frozen soils for cold regions science and engineering.

Comparative Analysis of Machine Learning Techniques and Data Sources for Dead Tree Detection: What Is the Best Way to Go?

In Central Europe, the extent of bark beetle infestation in spruce stands due to prolonged high temperatures and drought has created large areas of dead trees, which are difficult to monitor by ground surveys. Remote sensing is the only possibility for the assessment of the extent of the dead tree areas. Several options exist for mapping individual dead trees, including different sources and different processing techniques. Satellite images, aerial images, and images from UAVs can be used as sources. Machine and deep learning techniques are included in the processing techniques, although models are often presented without proper realistic validation.This paper compares methods of monitoring dead tree areas using three data sources: multispectral aerial imagery, multispectral PlanetScope satellite imagery, and multispectral Sentinel-2 imagery, as well as two processing methods. The classification methods used are Random Forest (RF) and neural network (NN) in two modalities: pixel- and object-based. In total, 12 combinations are presented. The results were evaluated using two types of reference data: accuracy of model on validation data and accuracy on vector-format semi-automatic classification polygons created by a human evaluator, referred to as real Ground Truth. The aerial imagery was found to have the highest model accuracy, with the CNN model achieving up to 98% with object classification. A higher classification accuracy for satellite imagery was achieved by combining pixel classification and the RF model (87% accuracy for Sentinel-2). For PlanetScope Imagery, the best result was 89%, using a combination of CNN and object-based classifications. A comparison with the Ground Truth showed a decrease in the classification accuracy of the aerial imagery to 89% and the classification accuracy of the satellite imagery to around 70%. In conclusion, aerial imagery is the most effective tool for monitoring bark beetle calamity in terms of precision and accuracy, but satellite imagery has the advantage of fast availability and shorter data processing time, together with larger coverage areas.

A Survey on Information Bottleneck.

This survey is for the remembrance of one of the creators of the information bottleneck theory, Prof. Naftali Tishby, passing away at the age of 68 on August, 2021. Information bottleneck (IB), a novel information theoretic approach for pattern analysis and representation learning, has gained widespread popularity since its birth in 1999. It provides an elegant balance between data compression and information preservation, and improves its prediction or representation ability accordingly. This survey summarizes both the theoretical progress and practical applications on IB over the past 20-plus years, where its basic theory, optimization, extensive models and task-oriented algorithms are systematically explored. Existing IB methods are roughly divided into two parts: traditional and deep IB, where the former contains the IBs optimized by traditional machine learning analysis techniques without involving any neural networks, and the latter includes the IBs involving the interpretation, optimization and improvement of deep neural works (DNNs). Specifically, based on the technique taxonomy, traditional IBs are further classified into three categories: Basic, Informative and Propagating IB; While the deep IBs, based on the taxonomy of problem settings, contain Debate: Understanding DNNs with IB, Optimizing DNNs Using IB, and DNN-based IB methods. Furthermore, some potential issues deserving future research are discussed. This survey attempts to draw a more complete picture of IB, from which the subsequent studies can benefit.

An experimental analysis of machine learning techniques for crop recommendation

An experimental analysis of machine learning techniques for crop recommendation

Comparative analysis of different supervised methods for satellite-based land-use classification: a case study of Reyhanlı

Satellite-based land-use classification plays a crucial role in various Earth observation applications, ranging from environmental monitoring to disaster management. This study presents a comparative analysis of machine learning techniques applied to land cover classification using Landsat-9 and Sentinel-2 satellite imagery in the Reyhanlı district in southern Türkiye. Three different classification algorithms, Random Forest (RF), Support Vector Machine (SVM), and Maximum Likelihood Classification (MLC), were evaluated for their ability to distinguish different land cover classes. High resolution multispectral satellite imagery processed under the same conditions using Geographic Information System (GIS) software was utilized in this study. Visual inspection and statistical evaluation, including overall accuracy and kappa coefficient, were employed to assess classification performance. The classification of Sentinel-2 and Landsat-9 satellite imagery using different machine learning algorithms resulted in the highest overall accuracy (OA = 0.911, Kappa = 0.879) for Sentinel 2 imagery with the RF algorithm. These findings highlight the importance of satellite image selection and algorithm optimization for accurate land cover mapping. This study provides valuable insights for local planners and authorities and underscores the potential of Sentinel-2 imagery combined with machine learning techniques for effective land-use classification and monitoring.

Comparative Analysis of Machine Learning Techniques for Prediction of the Compressive Strength of Field Concrete

Comparative Analysis of Machine Learning Techniques for Prediction of the Compressive Strength of Field Concrete

A comparative analysis of machine learning techniques for building cooling load prediction

A comparative analysis of machine learning techniques for building cooling load prediction

Jucazinho Dam Streamflow Prediction: A Comparative Analysis of Machine Learning Techniques

Jucazinho Dam Streamflow Prediction: A Comparative Analysis of Machine Learning Techniques

Enhancing House Price Predictability: A Comprehensive Analysis of Machine Learning Techniques for Real Estate and Policy Decision-Making

Accurate house price prediction is crucial for stakeholders in real estate markets and economic policy formulation. This research investigates the application of sophisticated machine learning (ML) algorithms to improve the precision of house price forecasting. By analyzing existing literature, we explore the methodologies employed in house price prediction using ML approaches. We emphasize the significance of precise predictions for various stakeholders, including homebuyers, sellers, investors, and policymakers. Additionally, this abstract critically evaluates the strengths and limitations of different ML techniques in predicting housing prices Our goal is to enhance predictability of models through rigorous analysis, thus facilitating informed decision-making when it comes to housing transactions, investments, and policy implementations through our research.

Spontaneous brain activity in patients with central retinal artery occlusion: a resting-state functional MRI study using machine learning.

Central retinal artery occlusion (CRAO) is a serious eye condition that poses a risk to vision, resulting from the blockage of the central retinal artery. Because of the anatomical connection between the ocular artery, which derives from the internal carotid artery, and the anterior middle cerebral artery, hemodynamic alterations and sudden vision loss associated with CRAO may impact brain functionality. This study aimed to examine alterations in spontaneous neural activity among patients with CRAO by resting-state functional MRI. In addition, we selected the amplitude of low-frequency fluctuation (ALFF) and fractional amplitude of low-frequency fluctuation (fALFF) values as classification features for distinguishing CRAO from healthy controls (HCs) using a support vector machine classifier. A total of 18 patients diagnosed with CRAO and 18 HCs participated in the study. Resting-state brain function images and structural images were acquired from both groups. Aberrant changes in spontaneous brain functional activity among CRAO patients were investigated utilizing ALFF and fALFF analysis methods. Group differences in ALFF/fALFF values were assessed through a two-sample t-test. Subsequently, a machine learning classifier was developed to evaluate the clinical diagnostic potential of ALFF and fALFF values. In comparison to HCs, individuals with CRAO exhibited significantly higher ALFF values in the left cerebellum_6, vermis_7, left superior frontal gyrus, and left inferior frontal gyrus, triangular part. Conversely, the CRAO group displayed notably lower ALFF values in the left precuneus and left median cingulum gyri. Furthermore, higher fALFF values were observed in the left inferior frontal gyrus, triangular part, whereas lower fALFF values were noted in the right cerebellum_Crus2, left precuneus, right angular gyrus, left angular gyrus, right supramarginal gyrus, right superior parietal gyrus, and left precuneus. Utilizing the ALFF/fALFF values, the receiver operating characteristic curves (area under the curve) yielded 0.99 and 0.94 through machine learning analysis techniques. CRAO patients exhibit atypical neural activity in the brain, characterized by ALFF and fALFF values predominantly localized in the frontal, parietal, and cerebellar regions, which are closely linked to visual cognition and motor control impairments. Furthermore, ALFF and fALFF could serve as potential neuroimaging markers beyond the orbit among CRAO.