An adaptive global variable fidelity metamodeling strategy using a support vector regression based scaling function

Zhou, Qi; Shao, Xinyu; Jiang, Ping; Zhou, Hui; Shu, Leshi

doi:10.1016/j.simpat.2015.08.002

Cited by 75 publications

(17 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the insensitive loss function ε is adopted, the aim of SVR is to look for an f ( x ) which can make the difference between the true and the training values be less than the given error ε . Thus, function f ( x ) solution can be expressed as the following quadratic programming problem (Zhou et al ., ):

Φ (ω, ξ_{i}, ξ_{i}^{*}) = \frac{1}{2} {‖ ω ‖}^{2} + c \sum_{i = 1}^{normaln} (ξ_{i} + ξ_{i}^{*})

Subject to true{\begin{array}{l} y_{i} - (ω^{T} φ (x) + b) \leq ε + ξ_{i} \\ (ω^{T} φ (x) + b) - y_{i} \leq ε + ξ_{i}^{*} \\ (ξ_{i}, ξ_{i}^{*}) \geq 0 \end{array}

Where

ξ_{i}

and

ξ_{i}^{*}

are slack variables that determine the upper and lower excess deviations.

\frac{1}{2} {‖ ω ‖}^{2}

is the regularization term, c is the error penalty factor used to regulate the difference between the regularization term and empirical risk,

ε

is the loss function that determines the accuracy of the training data point.…”

Section: Methodsmentioning

confidence: 99%

Quantitative Determination of Rice Moisture Based on Hyperspectral Imaging Technology and BCC‐LS‐SVR Algorithm

Sun

Mao

et al. 2016

J Food Process Engineering

View full text Add to dashboard Cite

In this study, a method for quantitative determination of rice moisture based on hyperspectral imaging technology was proposed. First, the hyperspectral imaging system in the spectral range of 871-1766 nm was used to collect the hyperspectral images of 120 rice samples of 10 moisture grades. Support vector regression (SVR), least-squares support vector regression (LS-SVR), and bacterial colony chemotaxis least-squares support vector regression (BCC-LS-SVR) models were established to determine the moisture content by using full wavelengths spectra data. Among all the models, the BCC-LS-SVR model showed the best results. To simplify the calibration model, successive projections algorithm (SPA) was used for feature selection and the number of characteristic wavelengths was determined as 25. Principal component analysis (PCA) was used for feature extraction and the cumulative contribution rate of the first six principal components reached 99%, which could reflect most of the information of the full spectra data. Three new regression models based on the selected wavelengths were built and the results were improved obviously. The BCC-LS-SVR-SPA model got the best accuracy in prediction and calibration with R 2 p of 0.980, RMSEP of 0.967%, R 2 c of 0.985 and RMSEC of 0.591%. The overall results from this study demonstrated that hyperspectral image technology is feasible to detect rice moisture. PRACTICAL APPLICATIONSThe quality of rice has a direct relationship with the moisture content of rice. Because of the moisture content over standard, rice storage time becomes shorter and rice is easy to go bad. It's harmful to eat this rice for a long time. Traditional methods for identification of rice moisture mainly focus on the appearance of rice and depend on the feelings of professionals, which are tedious, time-consuming, expensive and greatly influenced by subjective factors. Hyperspectral imaging technology has the advantages of nondestructive, rapid, non-pollution, and so on. The results showed that hyperspectral imaging technology for the detection of the rice moisture is feasible and it can measure the moisture of rice.

show abstract

Φ (ω, ξ_{i}, ξ_{i}^{*}) = \frac{1}{2} {‖ ω ‖}^{2} + c \sum_{i = 1}^{normaln} (ξ_{i} + ξ_{i}^{*})

Subject to true{\begin{array}{l} y_{i} - (ω^{T} φ (x) + b) \leq ε + ξ_{i} \\ (ω^{T} φ (x) + b) - y_{i} \leq ε + ξ_{i}^{*} \\ (ξ_{i}, ξ_{i}^{*}) \geq 0 \end{array}

Where

ξ_{i}

and

ξ_{i}^{*}

are slack variables that determine the upper and lower excess deviations.

\frac{1}{2} {‖ ω ‖}^{2}

is the regularization term, c is the error penalty factor used to regulate the difference between the regularization term and empirical risk,

ε

is the loss function that determines the accuracy of the training data point.…”

Section: Methodsmentioning

confidence: 99%

Quantitative Determination of Rice Moisture Based on Hyperspectral Imaging Technology and BCC‐LS‐SVR Algorithm

Sun

Mao

et al. 2016

J Food Process Engineering

View full text Add to dashboard Cite

show abstract

“…144 It also has been shown within numerical analysis that the adaptive sampling methods yield superior surrogate approximation and lower computational expense compared to static techniques. 144 Researchers have reported adaptive sampling techniques for different surrogate model, such as support vector machines, 145,146 artificial neural network, 147 and others. [148][149][150] These are the type of adaptive sampling techniques where sampling points are placed systematically, yet still stochastically.…”

Section: Algorithmsmentioning

confidence: 99%

Automated robotic platforms in design and development of formulations

2021

View full text Add to dashboard Cite

Product design for formulations is an active and challenging area of research. The new challenges of a fast-paced market, products of increasing complexity, and practical translation of sustainability paradigms require to reexamine the existing theoretical frameworks to include the advantages deriving from the new reality of digitalization of business and research. In this work, we review the existing approaches, clearly stating the role of automation and machine-learning-guided optimization in the broader framework. Moving from this, we review the state of the art of automated hardware and software for formulated product design, and identify the open challenges for future research. Perspectives are given on the emerging fields of automated discovery, scale-up, and multistage optimization, and a unitary picture of the existing connections is provided, in the general context of a completely digital R&D workflow.

show abstract

“…Therefore, many supervised learning algorithms can be effectively used to develop simulation metamodels. These include neural networks (Alam, McNaught, & Ringrose, 2004; Can & Heavey, 2012; Kuo, Yang, Peters, & Chang, 2007; Xanthopoulos & Koulouriotis, 2018), kriging/Gaussian process (Dancik, Jones, & Dorman, 2010; Dosi, Pereira, & Virgillito, 2018; Kleijnen, 2009; Salle & Yıldızoğlu, 2014), SVR (Clarke, Griebsch, & Simpson, 2005; Edali & Yücel, 2018; Fonoberova, Fonoberov, & Mezić, 2013; Zhou, Shao, Jiang, Zhou, & Shu, 2015), RFs (Edali & Yücel, 2019; Villa‐Vialaneix, Follador, Ratto, & Leip, 2012), multivariate adaptive regression splines (Bozağaç, Batmaz, & Oğuztüzün, 2016; Friedman, 1991), radial basis functions (Hussain, Barton, & Joshi, 2002; Jakobsson, Patriksson, Rudholm, & Wojciechowski, 2010; Mullur & Messac, 2006) and first‐ and second‐order linear regression models (Durieux & Pierreval, 2004; Grow, 2017; Happe, Kellermann, & Balmann, 2006; Kleijnen & Deflandre, 2006). There are several studies in the literature which compare different subsets of these techniques based on different criteria such as accuracy, robustness, interpretability and efficiency (i.e., runtime) (Clarke, Griebsch, & Simpson, 2005; Li, Ng, Xie, & Goh, 2010; Østergård, Jensen, & Maagaard, 2018; Van Gelder, Das, Janssen, & Roels, 2014; Villa‐Vialaneix, Follador, Ratto, & Leip, 2012).…”

Section: Proposed Approachmentioning

confidence: 99%

Analysis of an individual‐based influenza epidemic model using random forest metamodels and adaptive sequential sampling

Edali

Yücel

2020

Syst Res Behav Sci

View full text Add to dashboard Cite

This study proposes a three‐step procedure for the analysis of input–response relationships of dynamic models, which enables the analyst to develop a better understanding about the dynamics of the system. The main building block of the procedure is a random forest metamodel capturing the input–output relationships. We utilize an active learning approach as the second step to improve the accuracy of the metamodel. In the last step, we develop a novel way to present the information captured by the metamodel as a set of intelligible IF–THEN rules. For illustration, we use the FluTE model, which is an individual‐based influenza epidemic model. We observe that the number of daily applicable vaccines determines the success of an intervention strategy the most. Another critical observation is that when the daily available vaccines are constrained, nonpharmaceutical strategies should be incorporated to reduce the extent of the outbreak.

show abstract

An adaptive global variable fidelity metamodeling strategy using a support vector regression based scaling function

Cited by 75 publications

References 45 publications

Quantitative Determination of Rice Moisture Based on Hyperspectral Imaging Technology and BCC‐LS‐SVR Algorithm

Quantitative Determination of Rice Moisture Based on Hyperspectral Imaging Technology and BCC‐LS‐SVR Algorithm

Automated robotic platforms in design and development of formulations

Analysis of an individual‐based influenza epidemic model using random forest metamodels and adaptive sequential sampling

Contact Info

Product

Resources

About