Study on Driving Decision-Making Mechanism of Autonomous Vehicle Based on an Optimized Support Vector Machine Regression

Modeling the behavior of groundwater levels is necessary to implement sustainable groundwater resource management. Groundwater is a non-linear and complex system, which can be modeled by data-driven models. This study evaluates the performances of data-driven models, support vector machine regression (SVR) and artificial neural network (ANN), for forecasting groundwater levels of confined and unconfined systems at 1-, 2-, and 3-month ahead. This is the first time that confined and unconfined aquifers have been compared using data-driven models. In addition, to identify the optimal input combination, a hybrid gamma test (GT) and genetic algorithm (GA) was used. The coefficient of correlation (R), Mean Absolute Error (MAE), root mean squared error (RMSE), Nash–Sutcliffe efficiency (NSE), and developed discrepancy ratio (DDR) were applied to evaluate the SVR and ANN models. Results showed that the SVR and ANN models were more accurate for the unconfined system than the confined system for forecasts up to 3-month ahead. In both hydrogeological systems for 1-month ahead, the models performed better than for 2- and 3-month ahead forecasts, and the accuracy of the models decreased as the months ahead increased. The SVR model performed better than the ANN model for 1-, 2-, and 3-month ahead groundwater-level forecasting. The SVR model could be successfully used in predicting monthly groundwater in confined and unconfined systems.

Application of robust principal component analysis–multivariate adaptive regression splines for the determination of °API gravity in crude oil samples using ATR-FTIR spectroscopy

Gas–solid fluidized beds are complex particle systems, and the electrostatic behavior of particles in fluidized beds is even more complex, which is influenced by numerous factors such as particle properties and operating conditions. Current studies focus on the effect of a certain factor on particle charging without a global picture. Furthermore, there is no mathematical model that can describe the interaction of multiple factors on particle charging because it is difficult to build a model for such a complex system. Therefore, a new approach is needed. In this study, a model capable of accurately predicting the surface charge density of particles in monodispersed gas–solid fluidized beds within a certain range was developed based on the literature and experimental data through several machine learning methods including kernel ridge regression (KRR), support vector machine regression (SVR), and multilayer perceptron (MLP). SVR and MLP models gave the best results with R2 equal to 0.980 and 0.979, respectively. However, the sensitivity analysis showed that the MLP model was more reliable than the SVR model. In conclusion, the feasibility of using machine learning to analyze the charging behavior of particles in fluidized beds is demonstrated, and the proposed MLP model can serve as an accurate correlative tool for fast and effective estimation of particle surface charge density in gas–solid fluidized beds.

Empirical approach for strength prediction of geopolymer stabilized clayey soil using support vector machines

Support vector machine regression (SVR) technique was used for predicting the uniaxial compressive strength of FRP confined concrete. Two different SVR models, viz. SVR model for carbon fiber reinforced polymer confined concrete and SVR model for glass fiber reinforced polymer confined concrete, were developed. The prediction efficacy of developed SVR models was compared with that of ANN and existing empirical FRP strength prediction models. Furthermore, for assessing variable contributions in strength prediction process, a sensitivity analysis is also presented. The study showed that SVR could be a powerful alternative physical tool for strength prediction of FRP confined concrete.

In order to quantitatively predictive the content of the coalbed methane (CBM), we make use of the known parameters of the core tests data to establish the support vector machine regression model between the core data and coal-bed methane content. The model is based on the small sample size theory. Using the model, we can predict the volume of gas content. We choose the coal seam thickness, coal vitrinite reflectance value and coal ash 3 parameters as input feature vectors, and coal-bed methane content as the output vector of support vector machine regression prediction model. Application of the proposed model in Binchang mining shows that the prediction error between the measured results and prediction are small and meet the accuracy requirements.

This study aimed to compare stepwise multiple linear regression (SMLR), partial least squares regression (PLSR) and support vector machine regression (SVMR) for estimating soil total nitrogen (TN) contents with laboratory visible/near-infrared reflectance (Vis/NIR) of selected coarse and heterogeneous soils. Moreover, the effects of the first (1st) vs. second (2nd) derivative of spectral reflectance and the importance wavelengths were explored. The TN contents and the Vis/NIR were measured in the laboratory. Several methods were employed for Vis/NIR data pre-processing. The SMLR, PLSR and SVMR models were calibrated and validated using independent datasets. Results showed that the SVMR and the PLSR models had similar performances, and better performances than the SMLR. The spectral bands near 1450, 1850, 2250, 2330 and 2430 nm in the PLSR model were important wavelengths. In addition, the 1st derivative was more appropriate than the 2nd derivative for spectral data pre-processing. PLSR was the most suitable method for estimating TN contents in this study. SVMR may be a promising technique, and its potential needs to be further explored. Moreover, the future studies using outdoor and airborne/satellite hyperspectral data for estimating TN content are necessary for testing the findings.

Choosing appropriate multivariate calibration and preprocessing transformation techniques is important in the determination of soil organic carbon (SOC) content based on visible and near-infrared (Vis–NIR) spectroscopy. The performance levels of partial least-squares regression (PLSR), support vector machine regression (SVMR), and wavelet neural network (WNN) calibration methods coupled with different preprocessing approaches were compared using three kinds of criteria, including the coefficient of determination (R2), root mean square error (RMSE), and residual prediction deviation (RPD). A total of 328 soil samples collected from the south bank of Hangzhou Bay were used as the dataset for the calibration–validation procedure and SOC content inversion. The effects of spectra preprocessing transformation methods were evaluated for raw spectra, Savitzky–Golay smoothing with the first derivatives of reflectance (FDR) and Savitzky–Golay smoothing with logarithm of reciprocal of the reflectance (log R−1). The results indicate that the SVMR is superior to the PLSR, and WNN models for SOC content prediction. The combination of the SVMR model with FDR provided the best prediction results for the SOC content, with R2p = 0.92, RPDP = 2.82, RMSEP = 0.36%, and a kappa correlation coefficient of interpolation as high as 0.97. The FDR of Vis–NIR spectroscopy combined with the SVMR model is recommended over the PLSR and WNN modeling techniques for the high-accuracy determination of the SOC content.

Development of a screening method for adulteration detection in coconut milk via mid-infrared spectroscopy: A study of linear and nonlinear regression method

Towards an effective cross-task mental workload recognition model using electroencephalography based on feature selection and support vector machine regression

In this paper, a robust image watermarking technique has been proposed based on the combination of discrete wavelet transform (DWT) and discrete cosine transform (DCT). SVM regression model has been incorporated for geometric distortion correction to achieve improved robustness against de-synchronization attacks such as rotation, translation etc. Low order Pseudo Zernike (PZ) moments have used as a feature vector in SVM regression model. The 3-level DWT transformed DCT coefficients are modified to embed a binary bit. The performance of algorithm has been observed against both intentional and non-intentional attacks. The scheme provides an average imperceptibility of around 42.45 dB. The effect of rotation and translation attack has been estimated using trained SVM regression model. The robustness against de-synchronization attack is performed after correcting the attacked watermarked images. The experimental results show that the algorithm provides adequate robustness against both the geometric and non-geometric attacks.

In order to effectively detect the distribution of Cd in lettuce leaves, this article proposes a wavelet support vector machine regression (WSVR) modeling method and uses it in the prediction of cadmium (Cd) content in lettuce leaves. In addition, this article also observed the internal and external tissue structure of lettuce leaves under different concentrations of Cd stress by scanning electron microscopy and transmission electron microscopy. Moreover, the lettuce leaf samples under different Cd concentration stresses were acquired by the hyperspectral imaging system, different preprocessing algorithms combined with dimensionality reduction algorithms were used to select the spectral characteristic wavelengths, and SVR and WSVR models were established based on the spectral characteristic data. The results showed that with the increase of Cd concentration, the internal and external tissue structure of lettuce leaves changed significantly. The WSVR model parameters for the best prediction of Cd concentration in lettuce were Rp2 of 0.8843, RMSEP of 0.1292 mg/kg. Finally, the Cd contents in lettuce leaves were visualized on the prediction maps by predicted spectral features on each hyperspectral image pixel.Practical ApplicationsIt is of great significance to realize the visual expression of the distribution of heavy metals in crop leaves through nondestructive testing. In order to demonstrate the feasibility of Vis‐NIR hyperspectral imaging technology for the detection of cadmium (Cd) content in lettuce leaves, scanning electron microscopy and transmission electron microscopy were used to observe the internal and external tissue structure of lettuce leaves under different concentrations of Cd. Wavelet support vector machine regression modeling method was proposed and used to predict the Cd content in lettuce leaves. It is proved that the Vis‐NIR hyperspectral imaging technology is a feasible and effective method to realize the visualization of Cd content distribution in crop leaves.

The monitoring and replacement of lubricating oil has an important impact on mechanical equipment. In this study, based on the infrared spectroscopy monitoring method, an acid value index prediction model is established. The support vector machine regression method is used to quantitatively analyze the acid number of the oil sample, which verifies the stability and predictive ability of the quantitative prediction model, and we provide a theoretical basis and practical examples for the online monitoring of oil indicators. In addition, a support vector machine regression model is established by observing the changing law of the spectral absorption peak and selecting the dominant wavelength, and it is compared against the prediction algorithm of the long- and short-term memory network. By comparing the deviation relationship between the predicted value and the real chemical value, the feasibility of the infrared spectroscopy prediction model is verified. The experimental results show that the correlation coefficient between the predicted value of the model and the actual measured value reaches 0.98. This proves that the prediction effect of the prediction model based on the infrared spectrum data and the support vector machine regression method is better than that of the long- and short-term memory network trend prediction model, and the predicted results are reliable.

This paper integrated superiority from time series model and least square support vector machine regression model with data aggregation for traffic speed short term forecasting. Based on the results of traffic data variations analysis, the practicability that speed data can be aggregated to several periods was confirmed, and aggregated model can be developed to forecast the speed with auto regression (AR) model and support vector machine regression (SVR). Then the speed data in case study were integrated to 4 periods at the location of Remote Traffic Microwave Sensors (RTMS) 2047 on 2nd Ring Road Expressway in Beijing. Arguments with coefficients from AR models then act as the independent variables of LSSVR in aggregated model. Short term traffic speed was predicted by aggregated model, and the results indicated that taking advantages of time periods variation rule inside the aggregated model would help save the model running time cost under the premise of accuracy with better prediction ability than LSSVR in certain conditions.

The development of compounds that selectively modulate thyroid hormone action by serving as subtype-selective ligands of the thyroid hormone receptors (TRs) would be useful for clinical therapy. In the present work, quantitative structure-activity relationship (QSAR) models by adopting molecular descriptors to predict the TR binding activity were established based on a data set of TR ligands. The linear (multiple linear regression (MLR) and partial least squares regression (PLSR)) and nonlinear (support vector machine regression (SVR)) methods were employed to investigate the relationship between structural properties and binding activities. The proposed PLSR model was slightly superior to the MLR model and SVR model, as indicated by the reasonable statistical properties (TRβ: Rtr2 = 0.9594, Qte2 = 0.8091. TRα: Rtr2 = 0.9705, Qte2 = 0.8057). Additionally, molecular docking simulations were also performed to study the probable binding modes of the ligands and the TR subtype selectivity. The results indicate that substituents located in region A, region B, and region C and the orientation of these groups might result in the subtype selectivity based on the hydrogen bonding and electrostatic interactions. The derived QSAR models together with the molecular docking results have good potential in facilitating the discovery of novel TR ligands with improved activity and subtype selectivity.