Machine Learning Optimization Research Articles

Predicting disc cutter wear using two optimized machine learning techniques

Predicting disc cutter wear using two optimized machine learning techniques

Optimized machine learning algorithm for thyroid tumour type classification: A hybrid approach Random Forest, and intelligent optimization algorithms

Thyroid tumours are a common form of cancer, and accurate classification of their type is crucial for effective treatment planning. This research presents a hybrid approach for the classification of thyroid tumours based on their type. The proposed approach combines the use of advanced machine learning techniques with a comprehensive database of thyroid tumour samples. The database includes various features such as tumour size, shape, and texture, as well as patient-specific information. The hybrid approach aims to optimize the classification process by leveraging the diverse set of features and utilizing the power of machine learning algorithms. By harnessing the power of machine learning algorithms, this approach has the potential to revolutionize the field of thyroid tumour classification and significantly improve patient outcomes. The optimization strategy is Particle Swarm Optimization, refining the classification performance and ensuring optimal accuracy in identifying and categorizing four types of thyroid tumours. The utilization of advanced diagnostic tools and state-of-the-art Random forest classifier techniques in this approach marks a significant advancement in the field of thyroid tumour classification. Through the augmentation of the dataset and the pre-processing techniques employed, the hybrid classification system demonstrates enhanced accuracy and reliability in distinguishing between different types of thyroid tumours. This innovative approach not only provides a more comprehensive understanding of thyroid tumours but also paves the way for personalized and effective treatment strategies, ultimately improving patient care and outcomes.

Weight Optimization for missing data prediction of Landslide Susceptibility Mapping in Remote sensing Analysis

To keep track of changes to the Earth's surface, extensive time series of data from remote sensing using image processing is required. This research is motivated by the effectiveness of computational modelling techniques; however, the problem of missing data is multifaceted. When data at numerous a-periodic timestamps are absent during multi-temporal analysis, the issue becomes increasingly problematic. To make remote sensing time series analysis easier, weight optimised machine learning is used in this study to rebuild lost data. Keeping the causality restriction in mind, this method makes use of data from previous and subsequent timestamps. The architecture is based on an ensemble of numerous forecasting modules, built on the observed data in the time-series order. Dummy data is used to connect the forecasting modules, which were previously linked by the earlier half of the sequence. After that, iterative improvements are made to the dummy data to make it better fit the next segment of the sequence. On the basis of Landsat-7 TM-5 satellite imagery, the work has been proven to be accurate in forecasting missing images in normalised difference vegetation index time series. In a performance evaluation, the proposed forecasting model was shown to be effective.

Erratum: Genetic algorithm enhanced by machine learning in dynamic aperture optimization [Phys. Rev. Accel. Beams 21 , 054601 (2018)

Erratum: Genetic algorithm enhanced by machine learning in dynamic aperture optimization [Phys. Rev. Accel. Beams <b>21</b> , 054601 (2018)

Optimized Machine Learning Model for Fire Consequence Prediction

This article focuses on using machine learning to predict the distance at which a chemical storage tank fire reaches a specified thermal radiation intensity. DNV’s Process Hazard Analysis Software Tool (PHAST) is used to simulate different scenarios of tank leakage and to establish a database of tank accidents. Backpropagation (BP) neural networks, random forest models, and the optimized random forest model K-R are used for model training and consequence prediction. The regression performance of the models is evaluated using the mean squared error (MSE) and R2. The results indicate that the K-R regression prediction model outperforms the other two machine learning algorithms, accurately predicting the distance at which the thermal radiation intensity is reached after a tank fire. Compared with the simulation results, the model demonstrates higher accuracy in predicting the distance of tank fire consequences, proving the effectiveness of machine learning algorithms in predicting the range of consequences of tank storage area fire events.

Developing an ensembled machine learning model for predicting water quality index in Johor River Basin

Currently, the Water Quality Index (WQI) model becomes a widely used tool to evaluate surface water quality for agriculture, domestic and industrial. WQI is one of the simplest mathematical tools that can assist water operator in decision making in assessing the quality of water and it is widely used in the last years. The water quality analysis and prediction is conducted for Johor River Basin incorporating the upstream to downstream water quality monitoring station data of the river. In this research, the numerical method is first used to calculate the WQI and identify the classes for validating the prediction results. Then, two ensemble and optimized machine learning models including gradient boosting regression (GB) and random forest regression (RF) are employed to predict the WQI. The study area selected is the Johor River basin located in Johor, Peninsular Malaysia. The initial phase of this study involves analyzing all available data on parameters concerning the river, aiming to gain a comprehensive understanding of the overall water quality within the river basin. Through temporal analysis, it was determined that Mg, E. coli, SS, and DS emerge as critical factors affecting water quality in this river basin. Then, in terms of WQI calculation, feature importance method is used to identify the most important parameters that can be used to predict the WQI. Finally, an ensemble-based machine learning model is designed to predict the WQI using three parameters. Two ensemble ML approaches are chosen to predict the WQI in the study area and achieved a R2 of 0.86 for RF-based regression and 0.85 for GB-based ML technique. Finally, this research proves that using only the biochemical oxygen demand (BOD), the chemical oxygen demand (COD) and percentage of dissolved oxygen (DO%), the WQI can be predicted accurately and almost 96 times out of 100 sample, the water class can be predicted using GB ensembled ML algorithm. Moving forward, stakeholders may opt to integrate this research into their analyses, potentially yielding economic reliability and time savings.

Optimizing machine learning techniques and SHapley Additive exPlanations (SHAP) analysis for the compressive property of self-compacting concrete

This study examined the effectiveness of employing machine learning (ML) techniques to estimate the compressive strength (CS) of self-compacting concrete (SCC). Multiple techniques were utilized, such as a decision tree (DT), a random forest regressor (RFR), an AdaBoost regressor (AR), and a gradient boosting regressor (GBR). Additionally, the research investigated the impacts and correlations among input variables and the CS of SCC using the SHapley Additive exPlanations (SHAP) technique. A full dataset was created, including a single dependent variable and ten independents. Although GBR and DT methods proved effective, the research indicated that AR and RFR provided the most accurate predictions of SCC's CS. The AR and RFR models demonstrated superior performance compared to the DT and GBR models, as evident by their R2 values of 0.90 and 0.91, respectively, outperforming the DT and GBR models with R2 values of 0.85 and 0.88, respectively. The SHAP investigation revealed that the concentration of superplasticizer and cement in the mixture had an extensive impact on the CS of SCC. The findings of this study indicate that both AR and RFR methods had comparable predictive abilities in estimating SCC's CS.

Parallel Processing Impact on Random Forest Classifier Performance: A CIFAR-10 Dataset Study

Using the CIFAR-10 dataset, this research investigates how parallel processing affects the Random Forest method's machine learning performance. Accuracy and training time are highlighted in the study as critical performance indicators. Two cases were studied, one with and one without parallel processing. The results show the strong prediction powers of the Random Forest algorithm, which continues to analyze data in parallel while retaining a high accuracy of 97.50%. In addition, training times are notably shortened by parallelization, going from 0.6187 to 0.4753 seconds. The noted increase in time efficiency highlights the importance of parallelization in carrying out activities simultaneously, which enhances the training process's computational efficiency. These results provide important new information about how to optimize machine learning algorithms using parallel processing approaches.

Ataxic person prediction using feature optimized based on machine learning model

Ataxic gait monitoring and assessment of neurological disorders belong to important areas that are supported by digital signal processing methods and artificial intelligence (AI) techniques such as machine learning (ML) and deep learning (DL) techniques. This paper uses spatio-temporal data from Kinect sensor to optimize machine learning model to distinguish between ataxic and normal gait. Existing ML-based methodologies fails to establish feature correlation between different gait parameters; thus, exhibit very poor performance. Further, when data is imbalanced in nature the existing ML-based methodologies induces higher false positive. In addressing the research issues this paper introduces an extreme gradient boost (XGBoost)-based classifier and enhanced feature optimization (EFO) by modifying the standard cross validation (SCV) mechanism. Experiment outcome shows the proposed ataxic person identification model achieves very good result in comparison with existing ML-based and DL-based ataxic person identification methodologies.

Transformative role of machine learning in design optimization of reinforced concrete frames

Transformative role of machine learning in design optimization of reinforced concrete frames

Optimizing Machine Learning Models for Urinary Tract Infection Diagnostics: A Comparative Study of Logistic Regression and Random Forest

Urinary Tract Infections (UTIs) present a significant healthcare challenge due to their prevalence and diagnostic complexity. Timely and accurate diagnosis is critical for effective treatment, yet traditional methods like microbial cultures and urinalysis are often slow and inconsistent. This study introduces machine learning (ML) as a transformative solution for UTI diagnostics, particularly focusing on logistic regression and random forest models renowned for their interpretability and robustness. We conducted a meticulous hyperparameter tuning process using a rich dataset from a clinic in Northern Mindanao, Philippines, incorporating demographic, clinical, and urinalysis data. Our research outlines a detailed methodology for applying and refining these ML models to predict UTI outcomes accurately. Through comprehensive hyperparameter optimization, we enhanced the predictive performance, demonstrating a significant improvement over standard diagnostic practice. The findings reveal a clear superiority of the random forest model, achieving a top testing accuracy of 0.9814, compared to the best-performing logistic regression model's accuracy of 0.7626. This comparative analysis not only validates the efficacy of ML in medical diagnostics but also emphasizes the potential clinical impact of these models in real-world settings. The study contributes to the burgeoning literature on ML applications in healthcare by providing a blueprint for optimizing ML models for clinical use, particularly in diagnosing UTIs. It underscores the promise of ML in augmenting diagnostic precision, thereby potentially reducing the global healthcare burden associated with UTIs.

What if the Covid-19 Pandemic Never Happened Estimation of the Tourist Arrivals for 2020 Via Levenberg-Marquardt Optimization and K-Star (K ) Machine Learning Algorithms

Based on the probability theory, which is used to model uncertainty and randomness in real-world situations, the study aims to understand the impact of uncertain conditions, such as the Covid-19 pandemic, on the accuracy of the algorithms and the resulting losses to a country's tourism industry. The contribution of this paper to the international body of knowledge is twofold: firstly, it advances theoretical understanding of the use of probability theory in modelling real-world problems; and secondly, it offers a methodological approach for estimating tourist arrivals that accounts for the impact of extreme events. To achieve these aims, the Levenberg-Marquardt optimization was first applied to determine the optimal coefficients of the exponential function for estimating tourist arrivals from 1950 to 2020. Next, the K-Star machine learning algorithm was applied to the dataset with and without Covid-19 cases to estimate tourist arrivals.

Unveiling Marketing Potential: Harnessing Advanced Analytics and Machine Learning for Gold Membership Strategy Optimization in a Superstore

Unveiling Marketing Potential: Harnessing Advanced Analytics and Machine Learning for Gold Membership Strategy Optimization in a Superstore

A Comparative Study on Feature Extraction Techniques for the Discrimination of Frontotemporal Dementia and Alzheimer's Disease with Electroencephalography in Resting-State Adults.

Early-stage Alzheimer's disease (AD) and frontotemporal dementia (FTD) share similar symptoms, complicating their diagnosis and the development of specific treatment strategies. Our study evaluated multiple feature extraction techniques for identifying AD and FTD biomarkers from electroencephalographic (EEG) signals. We developed an optimised machine learning architecture that integrates sliding windowing, feature extraction, and supervised learning to distinguish between AD and FTD patients, as well as from healthy controls (HCs). Our model, with a 90% overlap for sliding windowing, SVD entropy for feature extraction, and K-Nearest Neighbors (KNN) for supervised learning, achieved a mean F1-score and accuracy of 93% and 91%, 92.5% and 93%, and 91.5% and 91% for discriminating AD and HC, FTD and HC, and AD and FTD, respectively. The feature importance array, an explainable AI feature, highlighted the brain lobes that contributed to identifying and distinguishing AD and FTD biomarkers. This research introduces a novel framework for detecting and discriminating AD and FTD using EEG signals, addressing the need for accurate early-stage diagnostics. Furthermore, a comparative evaluation of sliding windowing, multiple feature extraction, and machine learning methods on AD/FTD detection and discrimination is documented.

Machine Learning Optimization Techniques: A Survey, Classification, Challenges, and Future Research Issues

Machine Learning Optimization Techniques: A Survey, Classification, Challenges, and Future Research Issues

Prevention of Thermal Crack in Steel Slab Using Neural Networks Model to Predict Impact Absorption Energy

To prevent thermal cracks on steel slabs during the stacking, scarfing, and grinding processes after continuous casting, an optimized machine learning (ML) model to predict the impact absorption energy (EIA) of steel slabs is developed. A total of 1,421 experimental EIA data are collected from Charpy impact tests at slab temperatures 25 ≤ TS ≤ 400 °C, then ML models that considered 15 steel‐component variables and temperatures are developed. Four ML models are developed and their accuracies are compared. The optimized deep neural network algorithm predicts EIA most accurately with root mean squared error of 10.82 J and coefficient of determination (R2) of 0.992. Then the predicted EIA is used as the criterion to predict the occurrence of thermal cracks in slabs that are actually produced by a continuous casting process. At TS = 250 °C, cracking does not occur when the steel has predicted EIA &gt; 175 J. The use of the developed model to predict EIA can prevent the formation of thermal cracks in slabs produced by continuous casting, and enable optimization of the cooling method and scarfing methods that precede the next process after continuous casting of slabs.

Circuit-level Optimization for Machine Learning: Enhancing Efficiency and Performance in Neural Network Accelerators

Circuit-level Optimization for Machine Learning: Enhancing Efficiency and Performance in Neural Network Accelerators

Novel Hybrid Deep Optimized Machine Learning Model for Energy Efficient Cluster Selection in Unmanned Aerial Vehicle Networks

Unmanned Aerial Vehicles (UAVs) have grown into a more powerful type of data transmission due to this rapid progress of evolution of wireless communication technology. In addition, UAVs have been proven to be effective in a variety of applications, including intelligent transport, disaster risk management, surveillance, and environmental monitoring. When UAVs are deployed randomly, however, they can effectively accomplish challenging tasks because of the UAVs’ has low battery capacity, quick mobility, and dynamic in nature orientation. Due to this reason, a new technique must be designed for an optimal energy efficient UAV clustering as well as data routing protocols. In this work proposes a new hybrid model of Emperor penguin-based Generalized Approximate Reasoning Based Intelligent Control (EP-GARIC) cluster-based network topology. Furthermore, the optimal routing function is achieved by the proposed Artificial Jellyfish Optimization (AJO). The implementation of this research is carried out using Network Simulator (NS2). The simulation results displays the effective performance of the suggested approach in terms of reduced energy consumption, improved packet delivery ratio, reduced loss, and so on over compared to the conventional approaches

A Bayesian optimized machine learning approach for accurate state of charge estimation of lithium ion batteries used for electric vehicle application

In recent times, battery technology used in electric vehicles has drawn numerous researchers' attention. Monitoring the battery condition, particularly the State of charge, is required to ensure the battery's safe and reliable performance. Despite the fact that many SOC estimation methods have been proposed, further research is necessary to identify an approach that adapts versatile lithium-ion battery chemistries. In the recent study it has been demonstrated that Machine Learning approaches have better prediction accuracy compared to conventional methods. To maximize the performance of Machine learning models, it is imperative to select the optimal hyperparameters and employ appropriate input parameters. At present, researchers employ, established heuristics methods to select hyperparameters, which may involve manual tuning or exhaustive search techniques such as random search and grid search. These techniques make the models less accurate and inefficient. In this paper, a systematic, automated process for selecting hyperparameters with a Bayesian optimization algorithm is proposed. In addition, along with the battery parameters (voltage, current and temperature), vehicle velocity, road condition, motor characteristics and environmental conditions are used as the input parameters for accurate SOC prediction. The highly correlated input features are selected through the MRMR algorithm. The performance of six ML algorithms, namely SVM, ANN, GPR, Ensembler, linear regression and Decision Tree, is tested and validated with and without hyperparameter tuning for different data sets. The experimental results demonstrate that the hyperparameter-tuned model outperforms the standard model in estimating the State of charge (SOC). In addition, the model's estimation error is <1 % across a range of vehicle velocity, road condition, motor torque, battery voltage, current, and temperature.

AdaGCR: An improved method for optimizing machine learning training process

In contemporary machine learning, training datasets are typically divided into batches, and models are updated incrementally through batch iterations to save memory and reduce overfitting. However, determining the optimal hyperparameters like learning rate, batch size and number of epochs remains a challenge which often relying on empirical insights. This paper explores a novel method called Adaptive Gradient Conflict Rate (AdaGCR) to optimize the training process. It leverages the idea of gradient conflict rate, which reflects the models position within a batch model set and accordingly adjusts the global learning rate. This proposed method is tested by training a Deep Neural Network (DNN) model with MNIST dataset which represents simple tasks and a ResNet-18 model with CIFAR-10 dataset which represents more complicated tasks for solving real world problems. Experiments conducted on DNN demonstrates the proposed methods effectiveness in reducing overfitting and enhancing convergence, particularly with a well-suited initial learning rate. However, its applicability to more complex models like ResNet-18 may require further refinements, such as layer-specific learning rate adjustments. Future research should focus on fine-tuning AdaGCR and extending its utility across diverse machine learning models and tasks.