BGRF: A broad granular random forest algorithm

Fu, Xingyu; Chen, Yingyue; Yan, Jingru; Chen, Yumin; Xing, Feng

doi:10.3233/jifs-223960

Cited by 8 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Accuracy (%) Support vector regression (SVR) [29] 89.2924 Random forest (RF) [30] 84.5236 Extreme learning machine (ELM) [31] 84.1132 General regression neural network (GRNN) [4] 74.4805 Proposed technique 91.9802…”

Section: Methodsmentioning

confidence: 99%

“…A microwave imaging algorithm is introduced to generate 3D microwave images as shown in figure 10 To enhance the robustness and accuracy of the detection system, an ensemble learning method is introduced to deal better with the six-layer images at different positions in the z-direction. The architecture of ensemble learning is shown in figure 10(e), which comprises six base learners and one [30] is introduced as a metalearner to obtain the final prediction of BGL.…”

Section: Bgl Estimation With the Bgl Approach Of Ensemble Learningmentioning

confidence: 99%

See 1 more Smart Citation

Accurate blood glucose level monitoring using microwave imaging

Wang,

Xiao,

Pang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Painless, and non-invasive detection techniques are needed to replace finger-pick blood for diabetic. In this paper, a first-of-its-kind, noninvasive, and continuous blood glucose level (BGL) detection method based on microwave imaging is introduced. This method avoids the complex task of frequency choice for the design of electromagnetic sensor. A radar-based microwave imaging technology combined with improved very-deep super resolution (VDSR-BL) method is introduced to obtain high-resolution (HR) microwave images. After image super-resolution reconstruction by VDSR-BL, the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) of HR images reach 35.4461 dB and 0.9761, respectively. Then an ensemble learning strategy based on support vector regression (SVR) and random forest algorithms is proposed to identify HR microwave images for BGL estimation. The developed detection system has been verified on the medium under tests (MUT) with different glucose solution. The final detection results obtain the root mean squared error (RMSE) of 0.1394 mg/mL and the mean absolute relative difference (MARD) of 8.02%, which show well accuracy with clinical acceptance. Meanwhile, we also conducted human trials. The high correlation coefficient (R) of 0.9254 was achieved between the results of microwave imaging and invasive BGL. Together, these results show that microwave imaging offers a promising new approach for noninvasive BGL monitoring.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Bgl Estimation With the Bgl Approach Of Ensemble Learningmentioning

confidence: 99%

Accurate blood glucose level monitoring using microwave imaging

Wang,

Xiao,

Pang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…For the fault type C j and the trained n decision trees [ 37 ], the prediction result of the random forest can be written as follows:

where T ( i ) denotes the diagnostic result of the ith tree, ∑ T ( i ) = C j indicates the number of diagnostic faults of type C j in all decision trees, and arg max is the subscript corresponding to the maximum value in a set of values.…”

Section: Remote Fault Diagnosis Model Based On Rfmentioning

confidence: 99%

“…For the fault type C j and the trained n decision trees [37], the prediction result of the random forest can be written as follows:…”

Section: Rf Algorithmmentioning

confidence: 99%

Remote Fault Diagnosis for the Powertrain System of Fuel Cell Vehicles Based on Random Forest Optimized with a Genetic Algorithm

Quan,

Zhang,

Feng

2024

Sensors

View full text Add to dashboard Cite

To enhance the safety and reliability of fuel cell vehicles, a remote monitoring system based on 5th generation (5G) mobile networks and controller area networks (CANs) was designed, and a random forest (RF) algorithm for the fault diagnosis for eight typical malfunctions of its powertrain system was incorporated. Firstly, the information on the powertrain system was obtained through a 5G-based monitoring terminal, and the Alibaba Cloud IoT platform was utilized for data storage and remote monitoring. Secondly, a fault diagnosis model based on the RF algorithm was constructed for fault classification; its parameters were optimized with a genetic algorithm (GA), and it was applied on the Alibaba Cloud PAI platform. Finally, the performance of the proposed RF fault diagnosis model was evaluated by comparing it with three other classification models: random search conditioning, grid search conditioning, and Bayesian optimization. Results show that the model accuracy, F1 score, and kappa value of the optimized RF fault classification model are higher than the other three. The model achieves an F1 value of 97.77% in identifying multiple typical faults of the powertrain system, as validated by vehicle malfunction data. The method demonstrates the feasibility of remote monitoring and fault diagnosis for the powertrain system of fuel cell vehicles.

show abstract

Adaptive intuitionistic fuzzy neighborhood classifier

Yuzhang,

Jusheng

2023

Int. J. Mach. Learn. & Cyber.

View full text Add to dashboard Cite

BGRF: A broad granular random forest algorithm

Cited by 8 publications

References 35 publications

Accurate blood glucose level monitoring using microwave imaging

Accurate blood glucose level monitoring using microwave imaging

Remote Fault Diagnosis for the Powertrain System of Fuel Cell Vehicles Based on Random Forest Optimized with a Genetic Algorithm

Adaptive intuitionistic fuzzy neighborhood classifier

Contact Info

Product

Resources

About