Gaussian Process Regression for Predictive But Interpretable Machine Learning Models: An Example of Predicting Mental Workload across Tasks

Caywood, Matthew S.; Roberts, Daniel M.; Colombe, Jeffrey B.; Greenwald, Hal S.; Weiland, Monica Z.

doi:10.3389/fnhum.2016.00647

Cited by 47 publications

(31 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Non‐parametric approaches like GPR and SVM theoretically assume infinite parameters and capture all the uncertainties in the data. These uncertainties are not captured in a parametric approach like ANN (Caywood, Roberts, Colombe, Greenwald, & Weiland, ). Additionally, ANN is a black‐box approach; however, GPR and SVM can be used for exploring and exploiting the inherent function defining the variations (Schulz, Speekenbrink, & Krause, ).…”

Section: Resultsmentioning

confidence: 99%

Foam mat drying of papaya using microwaves: Machine learning modeling

Qadri

Osama

Srivastava

2020

J Food Process Engineering

View full text Add to dashboard Cite

The aim of this article is to study the microwave-assisted foam mat drying of papaya to form papaya powder. The process of foam mat drying of papaya using microwaves was modeled by machine learning approaches like artificial neural network (ANN), support vector regression (SVR), and Gaussian process regression (GPR). Effect of microwave power (480-640 W), inlet air temperature (40-50 C), and thickness of foam (2-4 mm) on the rate of drying were studied. The performance of the models was evaluated on the basis of different performance matrices including root mean square error (RMSE), coefficient of determination (R 2 ), model predictive error, and Chi-square (χ 2 ). The microwave heating of the papaya foam reduced the drying time manifold. All three machine learning approaches were able to predict the drying process efficiently.SVR showed the best performance (R 2 = 0.96; RMSE = 0.03) followed by GPR (R 2 = 0.92; RMSE = 0.04) and ANN (R 2 = 0.91; RMSE = 0.04). SVR-based model was simulated to predict the effect of power, temperature, and thickness on drying rate.Machine learning approaches can be efficiently used for modeling and microwaveassisted foam mat drying. SVR-based model proves to be a good alternative of ANN.Practical Applications: Foam mat drying is the method of dehydrating for heatsensitive and viscous materials which cannot be dried by other conventional methods. It is cost-effective, simple, and provides high product quality. The use of microwave-assisted drying decreases the drying time manifold. Modeling of the drying process is significant for its scale-up to industrial scale. Machine learning techniques have the capability of learning the hidden factors involved in the process and thus provide better predictions as compared with statistical regression methods. In this study, three machine learning methods (artificial neural network [ANN], support vector regression [SVR], and Gaussian process regression [GPR]) were compared for their efficiency in modeling the foam-mat drying of papaya pulp using microwaves.SVR showed the best performance as compared with ANN and GPR.

show abstract

Section: Resultsmentioning

confidence: 99%

Foam mat drying of papaya using microwaves: Machine learning modeling

Qadri

Osama

Srivastava

2020

J Food Process Engineering

View full text Add to dashboard Cite

show abstract

“…This enhances the kriging model's reliability even without using many hyperparameters [9]. Furthermore, kriging is robust to unexpected events such as feature sets inconsistency or sudden reduction in quality of data [6]. Finally, the proposed ordinary kriging model's performance on seizure detection exceeds those of some machine learning models that were explored in this work.…”

Section: Proposed Solution Of This Papermentioning

confidence: 88%

Real-Time Automatic Seizure Detection Using Ordinary Kriging Method in an Edge-IoMT Computing Paradigm

et al. 2020

View full text Add to dashboard Cite

Epilepsy is one of the leading neurological diseases in the world, affecting approximately 70 million of the world's population and often results in early mortality if not properly managed. The primary purpose of seizure detection is to reduce threat to life in the event of a seizure crisis. Previous efforts in the literature concentrate mostly on performance based on accuracy and other similar metrics. However, there is a short time lapse between the onset of a seizure attack and a potential injury that could claim the life of the patient. Therefore, there is the need for a more time-sensitive seizure detection model. We hereby propose a real-time seizure detection model in an edge computing paradigm using the ordinary kriging method, relying on the premise that the brain can be modeled as a three-dimensional spatial object, similar to a geographical panorama where kriging excels. Fractal dimensional features were extracted from patients' electroencephalogram (EEG) signals and then classified using the proposed ordinary kriging model. The proposed model achieves a training accuracy of 99.4% and a perfect sensitivity, specificity, precision and testing accuracy. Hardware implementation in an edge computing environment results in a mean detection latency of 0.85 s. To the best of the authors' knowledge, this is the first work that uses the kriging method for early detection of seizure.

show abstract

“…In contrast, GPR is a well‐studied statistical model. The similarity modeling approach is intuitive, and its results are easy to interpret . In treatment response prediction, the kernel function allows for the identification of known subjects with similar conditions.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning to Predict Anti–Tumor Necrosis Factor Drug Responses of Rheumatoid Arthritis Patients by Integrating Clinical and Genetic Markers

Guan

Zhang

Quang

et al. 2019

Arthritis & Rheumatology

102

View full text Add to dashboard Cite

Objective. Accurate prediction of treatment responses in rheumatoid arthritis (RA) patients can provide valuable information on effective drug selection. Anti-tumor necrosis factor (anti-TNF) drugs are an important second-line treatment after methotrexate, the classic first-line treatment for RA. However, patient heterogeneity hinders identification of predictive biomarkers and accurate modeling of anti-TNF drug responses. This study was undertaken to investigate the usefulness of machine learning to assist in developing predictive models for treatment response.Methods. Using data on patient demographics, baseline disease assessment, treatment, and single-nucleotide polymorphism (SNP) array from the Dialogue on Reverse Engineering Assessment and Methods (DREAM): Rheumatoid Arthritis Responder Challenge, we created a Gaussian process regression model to predict changes in the Disease Activity Score in 28 joints (DAS28) for the patients and to classify them into either the responder or the nonresponder group. This model was developed and cross-validated using data from 1,892 RA patients. It was evaluated using an independent data set from 680 patients. We examined the effectiveness of the similarity modeling and the contribution of individual features.Results. In the cross-validation tests, our method predicted changes in DAS28 (ΔDAS28), with a correlation coefficient of 0.405. It correctly classified responses from 78% of patients. In the independent test, this method achieved a Pearson's correlation coefficient of 0.393 in predicting ΔDAS28. Gaussian process regression effectively remapped the feature space and identified subpopulations that do not respond well to anti-TNF treatments. Genetic SNP biomarkers showed small contributions in the prediction when added to the clinical models. This was the bestperforming model in the DREAM Challenge. Conclusion. The model described here shows promise in guiding treatment decisions in clinical practice, based primarily on clinical profiles with additional genetic information.

show abstract

Gaussian Process Regression for Predictive But Interpretable Machine Learning Models: An Example of Predicting Mental Workload across Tasks

Cited by 47 publications

References 33 publications

Foam mat drying of papaya using microwaves: Machine learning modeling

Foam mat drying of papaya using microwaves: Machine learning modeling

Real-Time Automatic Seizure Detection Using Ordinary Kriging Method in an Edge-IoMT Computing Paradigm

Machine Learning to Predict Anti–Tumor Necrosis Factor Drug Responses of Rheumatoid Arthritis Patients by Integrating Clinical and Genetic Markers

Contact Info

Product

Resources

About