Support Vector Regression Technique Research Articles

Employing a support vector regression technique coupled with optimizers to estimate the compressive strength of reinforced concrete connections

Employing a support vector regression technique coupled with optimizers to estimate the compressive strength of reinforced concrete connections

Data-driven battery health state estimation based on charging and switching cabinets

Accurate evaluation of the battery’s state of health (SOH) in the charging and switching cabinet is crucial to ensure the battery operates safely and reliably, while also reducing maintenance costs for the battery system. A support vector regression technique, utilising the sparrow algorithm optimisation, is suggested to improve the precision of assessing the battery’s state of health in the charging and switching cabinet. This algorithm tackles the task of identifying parameters in the conventional model. Analysing the ageing dataset of lithium batteries is the initial stage to determine the health parameters that signal the battery’s health state. The support vector machine regression algorithm is employed to select the kernel function and penalty factor, which are fine-tuned using the sparrow optimisation technique. The data is utilised to create the SSA-SVR model. The battery health status of the charging and switching cabinet is assessed. The study shows that the enhanced support vector regression model can effectively monitor the status of lithium-ion batteries and achieve superior estimation results for different types of batteries.

Transfer force perception skills to robot‐assisted laminectomy via imitation learning from human demonstrations

AbstractA comparative study of two force perception skill learning approaches for robot‐assisted spinal surgery, the impedance model method and the imitation learning (IL) method, is presented. The impedance model method develops separate models for the surgeon and patient, incorporating spring‐damper and bone‐grinding models. Expert surgeons' feature parameters are collected and mapped using support vector regression and image navigation techniques. The imitation learning approach utilises long short‐term memory networks (LSTM) and addresses accurate data labelling challenges with custom models. Experimental results demonstrate skill recognition rates of 63.61%–74.62% for the impedance model approach, relying on manual feature extraction. Conversely, the imitation learning approach achieves a force perception recognition rate of 91.06%, outperforming the impedance model on curved bone surfaces. The findings demonstrate the potential of imitation learning to enhance skill acquisition in robot‐assisted spinal surgery by eliminating the laborious process of manual feature extraction.

Augmented support vector regression with an autoregressive process via an iterative procedure

The Support Vector Regression (SVR) technique can approximate intricate systems by addressing learning and estimation challenges within a reproducing kernel Hilbert space, devoid of reliance on specific parameter assumptions. However, when dealing with correlated data like time series, the SVR method often falls short in accounting for underlying temporal structures, leading to limited enhancements in prediction efficiency. We introduce an enhanced SVR method that considers temporal correlations (TemporalSVR) to overcome this constraint. Our proposed method extends kernel functions to include additional linear kernels, facilitating learning temporal patterns. Additionally, we develope an iterative training procedure for the augmented regression model. During model training, we estimate the hyper-parameter in the corresponding loss function using a ‘working’ likelihood approach, enhancing the generalization capabilities of the proposed regression. To demonstrate superior forecasting performance, we conduct extensive numerical simulations on both linear and nonlinear systems and the TemporalSVR achieves improvements ranging from 8% to 114% based on the RMSE ratio from the AR-X model. Furthermore, we investigate the forecasting performance of three basic models (NARX-NN, Statistical SVR, and SVR-ARIMA) and four deep learning (DL) techniques (Transformer, Informer, Reformer, Autormer, and Autoformer) by using a WTI forecasting study. Our proposed TemporalSVR achieves the smallest RMSE at 2.22 and attains the highest success ratio of stock direction prediction at 71.30%. All these numerical results highlight the effectiveness and advantages of our TemporalSVR in handling temporal data and making accurate predictions.

Machine learning aided uncertainty quantification for engineering structures involving material-geometric randomness and data imperfection

In real-world engineering, uncertainty is ubiquitous within material properties, structural geometry, load conditions, and the like. These uncertainties have substantial impacts on the estimation of structural performance. Furthermore, information or datasets in real life commonly contain imperfections, e.g., noise, outliers, or missing data. To quantify these impacts induced by uncertainties on structural behaviours and reduce the effects of data imperfections simultaneously, a machine learning-aided stochastic analysis framework is proposed. A novel supervised machine learning technique, namely the Capped Extended Support Vector Regression (CX-SVR) technique, is developed to effectively suppress the effects of outliers and noise in datasets. Its inherent convexity in optimization and capped strategy theoretically supports the accuracy of CX-SVR, especially in handling datasets with imperfections. Once the effective surrogate model is established, subsequent analyses, like sampling-based methods, can circumvent the cumbersome physical model, which is potentially the nest of computational burden and errors in engineering applications. The high robustness of the proposed approach can be summarized in four main aspects: unrestrictive selection of the system inputs and their statistical information, ‘perfect’ or ‘imperfect’ data, enough statistical information (including statistical moments, probability density functions, and cumulative distribution functions) of the system outputs, and physical problems from various engineering fields.

Machine Learning-Based Flood Forecasting System for Window Cliffs State Natural Area, Tennessee

The prevalence of unforeseen floods has heightened the need for more accurate flood simulation and forecasting models. Even though forecast stations are expanding across the United States, the coverage is usually limited to major rivers and urban areas. Most rural and sub-urban areas, including recreational areas such as the Window Cliffs State Natural Area, do not have such forecast stations and as such, are prone to the dire effects of unforeseen flooding. In this study, four machine learning model architectures were developed based on the long short-term memory, random forest, and support vector regression techniques to forecast water depths at the Window Cliffs State Natural Area, located within the Cane Creek watershed in Putnam County, Tennessee. Historic upstream and downstream water levels and absolute pressure were used to forecast the future water levels downstream of the Cane Creek watershed. The models were tested with lead times of 3, 4, 5, and 6 h, revealing that the model performances reduced with an increase in lead time. Even though the models yielded low errors of 0.063–0.368 ft MAE, there was an apparent delay in predicting the peak water depths. However, including rainfall data in the forecast showed a promising improvement in the models’ performance. Tests conducted on the Cumberland River in Tennessee showed a promising improvement in model performance when trained with larger data.

A new experimental design to predict carbon dioxide emissions using Boruta feature selection and hybrid support vector regression techniques

A new experimental design to predict carbon dioxide emissions using Boruta feature selection and hybrid support vector regression techniques

A new experimental design to predict carbon dioxide emissions using Boruta feature selection and hybrid support vector regression techniques

A new experimental design to predict carbon dioxide emissions using Boruta feature selection and hybrid support vector regression techniques

Development of machine learning-based solubility models for estimation of Hydrogen solubility in oil: Models assessment and validation

This research was done to build computational models for estimating solubility of hydrogen (S) in a given system based on the inputs of temperature (T) and pressure (P). In fact, multiples models were built considering double inputs and single output. To achieve this, three different regression algorithms were trained and tested on a comprehensive dataset using the optimization technique of Grey Wolf Optimization (GWO). The three models used were Elastic Net, Support Vector Regression (SVR), and Automatic Relevance Determination (ARD) Regression. R2 score, root mean square error (RMSE), and mean absolute error were just few of the measures used to assess the models' overall efficacy (MAE). The Elastic Net model achieved the highest R2 score of 0.965 with an RMSE of 8.84×10−2 and an MAE of 7.79×10−2. The SVR technique also performed well, indicated R2 of 0.949, an RMSE of 9.81×10−2, and an MAE of 7.79×10−2. The ARD Regression model revealed an R2 of 0.905 with an RMSE of 1.53×10−1 and an MAE of 1.32×10−1. These findings highlight the potential of machine learning models and metaheuristic algorithms such as GWO in accurately estimating gas solubility in oil, which can have significant implications for various industrial applications.

Hybrid Machine-Learning-Based Prediction Model for the Peak Dilation Angle of Rock Discontinuities.

The peak dilation angle is an important mechanical feature of rock discontinuities, which is significant in assessing the mechanical behaviour of rock masses. Previous studies have shown that the efficiency and accuracy of traditional experimental methods and analytical models in determining the shear dilation angle are not completely satisfactory. Machine learning methods are popular due to their efficient prediction of outcomes for multiple influencing factors. In this paper, a novel hybrid machine learning model is proposed for predicting the peak dilation angle. The model incorporates support vector regression (SVR) techniques as the primary prediction tools, augmented with the grid search optimization algorithm to enhance prediction performance and optimize hyperparameters. The proposed model was employed on eighty-nine datasets with six input variables encompassing morphology and mechanical property parameters. Comparative analysis is conducted between the proposed model, the original SVR model, and existing analytical models. The results show that the proposed model surpasses both the original SVR model and analytical models, with a coefficient of determination (R2) of 0.917 and a mean absolute percentage error (MAPE) of 4.5%. Additionally, the study also reveals that normal stress is the most influential mechanical property parameter affecting the peak dilation angle. Consequently, the proposed model was shown to be effective in predicting the peak dilation angle of rock discontinuities.

Optimizing Bio-Sensor Design With Support Vector Regression Technique for AlGaN/GaN MOS–HEMT

Optimizing Bio-Sensor Design With Support Vector Regression Technique for AlGaN/GaN MOS–HEMT

Improving Shallow Water Bathymetry Inversion through Nonlinear Transformation and Deep Convolutional Neural Networks

Nearshore bathymetry plays an essential role in various applications, and satellite-derived bathymetry (SDB) presents a promising approach due to its extensive coverage and comprehensive bathymetric map production capabilities. Nevertheless, existing retrieval techniques, encompassing physics-based and pixel-based statistical methodologies such as support vector regression (SVR), band ratio, and Kriging regression, exhibit limitations stemming from the intricate water reflectance process and the under-exploitation of the spatial component inherent in SDB. To surmount these obstacles, we introduce employment of deep convolutional networks (DCNs) for SDB in this study. We assembled multiple scenes utilizing networks with varying scale emphasis and an assortment of satellite datasets characterized by distinct spatial and spectral resolutions. Our findings reveal that these deep learning models yield high-caliber bathymetry outcomes, with nonlinear normalization further mitigating residuals in shallow water regions and substantially enhancing retrieval performance. A comparative analysis with the prevalent SVR technique substantiates the efficacy of the proposed methodology.

An Investigation of the Influence on Compacted Snow Hardness by Density, Temperature and Punch Head Velocity

The density, temperature, and punch head velocity are dominant factors to the variation of the compacted snow hardness measured by penetrometers. This effect is essential to the construction and operation of compacted snow roads. The Improved Motor-driven Snow Penetrometers (IMSP) are utilized in this research to control the penetration speed and measure the true cone hardness during snow penetration. This study employs a multi-method approach combining orthogonal experiments and the Support Vector Regression (SVR) technique to analyze the effects of these three factors on snow hardness. The results of this investigation indicate that, under identical conditions, density is positively correlated with the hardness of compacted snow, and its sensitivity and significance to the compacted snow hardness are the greatest. Temperature and penetration speed have an effect on hardness, which cannot be completely ignored. The hardness of snow close to its melting point, regardless of its density, decreases significantly at high penetration rates. This study investigates the factors that influence the hardness of compacted snow and provides substantial technical support for the design, construction, and maintenance of snow roads.

Economic Sustainability of Building and Construction Projects Based on Artificial Intelligence Techniques

Artificial intelligence (AI) has been shown to be an effective replacement for conventional modelling approaches. AI is a subfield of computer science that develops software and tools that mimic human intelligence. AI offers advantages over traditional methods for handling ambiguous circumstances. In addition, AI-based solutions can successfully replace testing when identifying engineering design parameters, saving a lot of time and resources. AI can also increase computer efficiency, decrease mistake rates, and speed up decision-making. Recently, there has been a lot of interest in machine learning (ML), a new area of cutting-edge intelligent methods for use in structural engineering. Consequently, this work presents a study on the economic management of building and construction projects based on creating ML techniques. It begins with an overview of the value of applying AI techniques in building and construction industry. The analysis of the prediction of reinforced concrete’s compressive strength while taking cost into account is then done using empirical data based on a case study. Accordingly, the findings showed that the support vector regression (SVR) and k-Nearest Neighbour (KNN) intelligence techniques are helpful in the construction business for controlling the strength of concrete based on sustainable cost reduction.

Enhanced extraction of brominated flame retardants from e-waste plastics

Acrylonitrile butadiene styrene (ABS) is a high utility engineering thermoplastic, abundantly present in waste electronic and electrical goods. The demand for this plastic resin increases continuously with rapid industrial growth across the world. During the recycling of e-waste plastics, it is a common occurrence to find brominated flame retardants, such as Tetrabromobisphenol A (TBBPA) in the plastic. These flame retardants are intentionally added to the plastics as additives. Three processes were investigated to find the optimum solvent/antisolvent combinations, process parameters for the efficient recovery of TBBPA free plastics: microwave assisted extraction (MAE), ultrasound assisted extraction (UAE), and dissolution reprecipitation (DR). Rigorous solvent screening was carried out for both TBBPA and ABS using Hansen Solubility Parameters (HSP), and their interactions with the solvents were investigated with a molecular based Conductor like Screening Model for Realistic Solvents (COSMO-RS) software platform. A 4:1 (v/v) solvent mixture of 1-propanol and heptane was found to be ideal for the solid–liquid extraction process. The optimal extraction conditions (Temperature = 110 – 130 °C, time = 60 to 70 min solid/liquid ratio of 0.025) in MAE were predicted using machine learning based Support Vector Regression (SVR) technique. MAE was found to be superior to UAE and DR processes in terms of quantity and quality of recovered plastics. The maximum extraction efficiencies for TBBPA removal were achieved with MAE, UAE, and DR processes at 95.81%, 93.92%, and 98.15%, respectively. However, the DR process had a relatively lower plastic recovery.

Prediction of Deposition Layer Morphology Dimensions Based on PSO-SVR for Laser–arc Hybrid Additive Manufacturing

Laser–arc composite additive manufacturing holds significant potential for a wide range of industrial applications, and the control of morphological dimensions in the deposited layer is a critical aspect of this technology. The width and height dimensions within the deposited layer of laser–arc hybrid additive manufacturing serve as essential indicators of its morphological characteristics, directly influencing the shape quality of the deposited layer. Accurate prediction of the shape dimensions becomes crucial in providing effective guidance for size control. To achieve precise prediction of shape dimensions in laser–arc composite additive manufacturing and ensure effective regulation of the deposited layer’s shape quality, this study introduces a novel approach that combines a particle swarm algorithm (PSO) with an optimized support vector regression (SVR) technique. By optimizing the SVR parameters through the PSO algorithm, the SVR model is enhanced and fine-tuned to accurately predict the shape dimensions of the deposited layers. In this study, a series of 25 laser–arc hybrid additive manufacturing experiments were conducted to compare different approaches. Specifically, the SVR model was built using selected radial basis function (rbf) kernel functions. Furthermore, the penalty factors and kernel parameters of the SVR model were optimized using the particle swarm optimization (PSO) algorithm, leading to the development of a PSO-SVR prediction model for the morphological dimensions of the deposited layers. The performance of the PSO-SVR model was compared with that of the SVR, BPNN, and LightGBM models. Model accuracy was evaluated using a test set, revealing average relative errors of 2.39%, 7.719%, 9.46%, and 5.356% for the PSO-SVR, SVR, BPNN, and LightGBM models, respectively. The PSO-SVR model exhibited excellent prediction accuracy with minimal fluctuations in prediction error. This performance demonstrates the model’s ability to effectively capture the intricate and non-linear relationship between process parameters and deposition layer dimensions. Consequently, the PSO-SVR model can provide a foundation for the control of morphological dimensions in the deposition layer, offering an effective guide for deposition layer morphology dimension control in laser–arc composite additive manufacturing.

AN ESTIMATION OF NUMBER OF DAILY DEATHS DUE TO COVID-19 IN UNITED STATES OF AMERICA, UNITED KINGDOM AND TURKEY

Covid-19 virus is threatening the world with health, social and economic implications and all around the world data is obtained continuously with pandemic for modelling and predicting the future. In this work, support vector regression technique was used to make some predictions on the daily death values due to Covid-19 virus. The models were created for the world, United States of America, United Kingdom and Turkey. All the regression models were tested using coefficient of determination (R2) and root mean square error (RMSE) values. The analysis was also conducted for comparing the suitability of linear, radial and polynomial kernels. The radial kernel produced relatively better results. In predicting the world data support vector regression with radial kernel produced 0.805262 R2 value on test data. In the models created for United States of America 0.723376 R2 value, for United Kingdom 0.95412 R2 value and for Turkey 0.875343 R2 value using test data were observed. Also, while the models were created for specific countries the comparisons were made between using only data for the country and also using the whole world data. In general modelling using the data for the world combined with the country data gave better prediction.

LinVec: A Stacked Ensemble Machine Learning Architecture for Analysis and Forecasting of Time-Series Data

Objectives: The proposed work integrates multiple Machine Learning approaches in a single model to be used for the analysis and forecasting of Time Series data. Methods: In the present work, the concept of Stacked Ensemble learning is proposed that uses Multiple Linear Regression and Support Vector Regression techniques as the base models. A Meta Model is constructed based on Multiple Linear Regression with necessary modifications. The outputs from the base models are fed into the meta-model which is mended with the capability of combining the predictions from the two base models to produce better results than the individual constituent parts, after running a k-fold training procedure. Findings: The proposed model is capable of analyzing and predicting any Time Series data. In the present study, stock data of six companies enlisted in the National Stock Exchange of India are analyzed for the prediction of the next day’s Open, High, and Low prices. The proposed work achieves better accuracy and reduces the error in prediction when compared to similar works done in the same field. Novelty: The amalgamated technique used in this work can be considered as a generalization of the stacked ensemble method in a broader aspect. The proposed model combines the strengths of multiple Machine Learning methods into a single model to achieve better performance than its individual counterparts. Further, several recent works have tried to predict only the next day’s Open and Closing Prices of stocks, but for an intraday trader, prediction of the next day’s Low and High prices of a stock are more significant than the closing prices. Very few works have predicted all of the Open, High and Low prices in a single study, our present work achieved this quite successfully. Keywords: Machine Learning; Stacked Ensemble Model; Support Vector Regression; Multiple Linear Regression; Time Series data analysis; Stock Price Prediction

A hybrid non‐parametric ground motion model for shallow crustal earthquakes in Europe

AbstractIn the current study, ground motion models (GMMs) are derived using the European Strong Motion (ESM) database for pseudo‐spectral acceleration (PSA), peak ground acceleration (PGA), peak ground velocity (PGV), peak ground displacement (PGD), cumulative absolute velocity (CAV), arias intensity (Ia), and significant duration. In addition to addressing random effects associated with ground motion regression, such as inter‐event, inter‐site, inter‐locality, and inter‐region variabilities, the current study also aims at reducing the standard deviations (STDs) of the GMMs through development of a hybrid non‐parametric GMM. The hybrid model is derived through an ensemble‐weighted method of five non‐parametric machine learning models: shallow neural network, deep neural network (DNN), gated recurrent unit (GRU), support vector, and random forest (RF) regression techniques; with weights based on model performances. The resulting hybrid model, which also accounts for epistemic uncertainty, is compared against other regional models and is found superior for all output variables. The inter‐event, inter‐site, inter‐locality, and inter‐region deviations, and total ergodic sigma of PSA for the ensemble model lies between 0.3164–0.4478, 0.4156–0.5339, 0.1449–0.3687, 0.0819–0.2421, and 0.668–0.8545, respectively. The coefficient of determination (R2) between predicted and recorded values lies between 0.8435–0.9114 for all the output variables.

Wafer-level packaging solder joint reliability lifecycle prediction using SVR-based machine learning algorithm

Abstract The development of new electronic packaging structures often involves a design-on-simulation approach. However, simulation results can be subjective, and there can be variances in outcomes depending on who is conducting the simulation. To address this issue, packaging designers are now turning to machine learning to increase the accuracy and efficiency of the design process. This research study focuses on using support vector regression (SVR) techniques, such as single kernel, multiple kernels and a new SVR technique, to predict the reliability of the wafer-level packaging (WLP). By doing so, the study aims to provide designers with a reliable way to assess the reliability life cycle of their packaging designs. This research includes three steps: validating the WLP's reliability using finite element analysis (FEA) and experiment results, using the validated FEA result as input to obtain a predictive model through the SVR technique and the evaluating predictive model's performance. The results show that the predictive models developed using the SVR technique have stable performance on different testing data, which is consistent with the FEA results.