Yang Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Yang Yu has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Zhongjie Wang
College of Electronics and Information Engineering, Tongji University, Shanghai, China
4 shared publications - Jinfu Fan
College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China
2 shared publications - Yuanqing Xian
Department of Control Science and Engineering, Tongji University, Shanghai, China
2 shared publications Chengchao Lu
Department of Control Science and Engineering, Tongji University, Shanghai 201804, P. R. China
1 shared publication- Chengchao Lu
Department of Control Science and Engineering, Tongji University, Shanghai, China
1 shared publication - Chuankui Yan
College of Mathematics and Physics, Wenzhou University, Wenzhou, China
1 shared publication
Latest Publications
- 28 April 2026
- Published by SAGE Publications in Technology and Health Care
A deep learning method for children's self-care problems classification using represent learning and focal loss
Background: An accurate diagnosis of children's self-care problems significantly matters in the growth and development of children. However, various and extensive disorders make the self-care problems classification extremely complex and require much effort and time to solve. Objective: To deal with the above challenge, a deep learning model is proposed to classify the children's self-care problems intelligently and precisely. Method: The proposed deep learning model contains two sub-deep neural networks. The first sub-network employs a technology of representing learning named triplet loss. It aims to compress the dimensions of the feature of the children with self-care problems to extract the useful information and exclude the noise, in order to improve classification performance. The second sub-network utilizes a technology for handling the class imbalance problem called focal loss to further improve the classification accuracy. Result: The experimental results show that the proposed deep learning model outperforms. The averages of accuracy, precision, recall, and F1 score can achieve 99.78%, 0.99, 0.99, and 0.99, respectively. Conclusion: To the best of our knowledge, the proposed method achieves state-of-the-art results. That can significantly support the rehabilitation and growth of children with self-care issues. Furthermore, this study also provides a demonstration and experience of the application of the deep learning model in the healthcare field.
- 28 April 2026
- Published by SAGE Publications in Technology and Health Care
A deep learning method for children's self-care problems classification using represent learning and focal loss
Background: An accurate diagnosis of children's self-care problems significantly matters in the growth and development of children. However, various and extensive disorders make the self-care problems classification extremely complex and require much effort and time to solve. Objective: To deal with the above challenge, a deep learning model is proposed to classify the children's self-care problems intelligently and precisely. Method: The proposed deep learning model contains two sub-deep neural networks. The first sub-network employs a technology of representing learning named triplet loss. It aims to compress the dimensions of the feature of the children with self-care problems to extract the useful information and exclude the noise, in order to improve classification performance. The second sub-network utilizes a technology for handling the class imbalance problem called focal loss to further improve the classification accuracy. Result: The experimental results show that the proposed deep learning model outperforms. The averages of accuracy, precision, recall, and F1 score can achieve 99.78%, 0.99, 0.99, and 0.99, respectively. Conclusion: To the best of our knowledge, the proposed method achieves state-of-the-art results. That can significantly support the rehabilitation and growth of children with self-care issues. Furthermore, this study also provides a demonstration and experience of the application of the deep learning model in the healthcare field.
- Open Access
- 12 April 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (8) , 3881
Graph Neural Network for Senior High Student’s Grade Prediction
Senior high school education (SHSE) forms a connecting link between the preceding junior high school education and the following college education. Through SHSE, a student not only completes k-12 education, but also lays a foundation for subsequent higher education. The grade of the student in SHSE plays a critical role in college application and admission. Therefore, utilizing the grade of the student as an indicator is a reasonable method to instruct and ensure the effect of SHSE. However, due to the complexity and nonlinearity of the grade prediction problem, it is hard to predict the grade accurately. In this paper, a novel grade prediction model aiming to handle the complexity and nonlinearity is proposed to accurately predict the grade of the senior high student. To deal with the complexity, a graph structure is employed to represent the students’ grades in all subjects. To handle the nonlinearity, the multi-layer perceptron (MLP) is used to learn (or fit) the inner relation of the subject grades. The proposed grade prediction model based on graph neural network is tested on the dataset of Ningbo Xiaoshi High School. The results show that the proposed method performs well in the prediction of senior high school student grades.
- 12 April 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (8) , 3881
Open AccessGraph Neural Network for Senior High Student’s Grade Prediction
Senior high school education (SHSE) forms a connecting link between the preceding junior high school education and the following college education. Through SHSE, a student not only completes k-12 education, but also lays a foundation for subsequent higher education. The grade of the student in SHSE plays a critical role in college application and admission. Therefore, utilizing the grade of the student as an indicator is a reasonable method to instruct and ensure the effect of SHSE. However, due to the complexity and nonlinearity of the grade prediction problem, it is hard to predict the grade accurately. In this paper, a novel grade prediction model aiming to handle the complexity and nonlinearity is proposed to accurately predict the grade of the senior high student. To deal with the complexity, a graph structure is employed to represent the students’ grades in all subjects. To handle the nonlinearity, the multi-layer perceptron (MLP) is used to learn (or fit) the inner relation of the subject grades. The proposed grade prediction model based on graph neural network is tested on the dataset of Ningbo Xiaoshi High School. The results show that the proposed method performs well in the prediction of senior high school student grades.
- 31 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 32 (8) , 4953-4967
Partial Label Learning Based on Disambiguation Correction Net With Graph Representation
Partial Label Learning (PLL) is a weakly supervised learning framework where each training instance is associated with more than one candidate label. This learning method is dedicated to finding out the true label for each training instance. Most of the current PLL algorithms directly disambiguate the candidate labels without correcting the disambiguated results, making the algorithms vulnerable to the influence of instances easily misjudged. In this paper, GraphDCN, an innovative disambiguation correction net with inductive graph representation learning model is proposed. GraphDCN consists of a disambiguation model and a correction model. For a given instance, the disambiguation model tries to fit its underlying ground-truth label through the candidate label distribution of the instances connected with the given one, while the correction model tries to maximize the distance between the disambiguated labels and non-candidate labels, and uses the label probability thresholds to correct the disambiguated labels that may be wrong. As the training goes on, both the disambiguation and correction models alternately and iteratively boost their performance. Moreover, when considering the implementation of the disambiguation model, a partial cross entropy formulation is proposed to estimate the ground-truth label loss by updating the ambiguity confidence matrix, which can be proved satisfying convergence in PLL. Simulation results reveal the overwhelming performance of GraphDCN.
- 31 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 32 (8) , 4953-4967
Partial Label Learning Based on Disambiguation Correction Net With Graph Representation
Partial Label Learning (PLL) is a weakly supervised learning framework where each training instance is associated with more than one candidate label. This learning method is dedicated to finding out the true label for each training instance. Most of the current PLL algorithms directly disambiguate the candidate labels without correcting the disambiguated results, making the algorithms vulnerable to the influence of instances easily misjudged. In this paper, GraphDCN, an innovative disambiguation correction net with inductive graph representation learning model is proposed. GraphDCN consists of a disambiguation model and a correction model. For a given instance, the disambiguation model tries to fit its underlying ground-truth label through the candidate label distribution of the instances connected with the given one, while the correction model tries to maximize the distance between the disambiguated labels and non-candidate labels, and uses the label probability thresholds to correct the disambiguated labels that may be wrong. As the training goes on, both the disambiguation and correction models alternately and iteratively boost their performance. Moreover, when considering the implementation of the disambiguation model, a partial cross entropy formulation is proposed to estimate the ground-truth label loss by updating the ambiguity confidence matrix, which can be proved satisfying convergence in PLL. Simulation results reveal the overwhelming performance of GraphDCN.
- Open Access
- 20 November 2021
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 16 (4) , 961-972
Detecting synaptic connections in neural systems using compressive sensing
Revealing synaptic connections between neurons is of great significance and practical value to biomedicine and bio-neurology. We present a general approach to reconstruct neuronal synapses, which is based on compressive sensing and special data processing. And this approach is more suitable for nervous system with peak time series. Numerical simulations illustrate the feasibility and effectiveness of the proposed approach. Moreover, this approach not only adapts to the asymmetry of neural connections and the diversity of coupling strength, but also adapts to the excitability and inhibition of neural node classification. In addition, the effects of the factors on the synaptic connection identification performance and their optimal states for the synaptic connection recovery are discussed. Besides, it is of great practical significance to control the order of Taylor expansion to improve the performance of synaptic connection recognition.
- 20 November 2021
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 16 (4) , 961-972
Open AccessDetecting synaptic connections in neural systems using compressive sensing
Revealing synaptic connections between neurons is of great significance and practical value to biomedicine and bio-neurology. We present a general approach to reconstruct neuronal synapses, which is based on compressive sensing and special data processing. And this approach is more suitable for nervous system with peak time series. Numerical simulations illustrate the feasibility and effectiveness of the proposed approach. Moreover, this approach not only adapts to the asymmetry of neural connections and the diversity of coupling strength, but also adapts to the excitability and inhibition of neural node classification. In addition, the effects of the factors on the synaptic connection identification performance and their optimal states for the synaptic connection recovery are discussed. Besides, it is of great practical significance to control the order of Taylor expansion to improve the performance of synaptic connection recognition.
- Open Access
YOT-Net: YOLOv3 Combined Triplet Loss Network for Copper Elbow Surface Defect Detection
Copper elbows are an important product in industry. They are used to connect pipes for transferring gas, oil, and liquids. Defective copper elbows can lead to serious industrial accidents. In this paper, a novel model named YOT-Net (YOLOv3 combined triplet loss network) is proposed to automatically detect defective copper elbows. To increase the defect detection accuracy, triplet loss function is employed in YOT-Net. The triplet loss function is introduced into the loss module of YOT-Net, which utilizes image similarity to enhance feature extraction ability. The proposed method of YOT-Net shows outstanding performance in copper elbow surface defect detection.
Open AccessYOT-Net: YOLOv3 Combined Triplet Loss Network for Copper Elbow Surface Defect Detection
Copper elbows are an important product in industry. They are used to connect pipes for transferring gas, oil, and liquids. Defective copper elbows can lead to serious industrial accidents. In this paper, a novel model named YOT-Net (YOLOv3 combined triplet loss network) is proposed to automatically detect defective copper elbows. To increase the defect detection accuracy, triplet loss function is employed in YOT-Net. The triplet loss function is introduced into the loss module of YOT-Net, which utilizes image similarity to enhance feature extraction ability. The proposed method of YOT-Net shows outstanding performance in copper elbow surface defect detection.
- 30 October 2018
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 09 (05)
Pseudo-derivative evolutionary algorithm and convergence analysis
In this paper, a novel evolutionary algorithm (EA), called pseudo-derivative EA (called PDEA), is proposed. The basic idea of PDEA is to use pseudo-derivative, which is obtained based on the information produced during the evolution, and to help search the solution of optimization problem. The pseudo-derivative drives the search process in a more informed direction. That makes PDEA different from the random optimization methods. The convergence of PDEA is first analyzed based on systems theory. The convergence condition of PDEA is then derived though this condition is too strong to be satisfied. Next, this condition is relaxed based on the entropy theory. Finally, performances of PDEA are evaluated on the benchmark functions and an adaptive liquid level control system of a surge tank. The numeric simulation results show that PDEA is capable of finding the solutions to the optimization problems with good accuracy, reliability, and speed.
- 30 October 2018
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 09 (05)
Pseudo-derivative evolutionary algorithm and convergence analysis
In this paper, a novel evolutionary algorithm (EA), called pseudo-derivative EA (called PDEA), is proposed. The basic idea of PDEA is to use pseudo-derivative, which is obtained based on the information produced during the evolution, and to help search the solution of optimization problem. The pseudo-derivative drives the search process in a more informed direction. That makes PDEA different from the random optimization methods. The convergence of PDEA is first analyzed based on systems theory. The convergence condition of PDEA is then derived though this condition is too strong to be satisfied. Next, this condition is relaxed based on the entropy theory. Finally, performances of PDEA are evaluated on the benchmark functions and an adaptive liquid level control system of a surge tank. The numeric simulation results show that PDEA is capable of finding the solutions to the optimization problems with good accuracy, reliability, and speed.
- 7 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 68 (1) , 87-94
A Joint Filter Approach for Reliable Power System State Estimation
Accurate power system state estimation facilitates power system control, optimization, and security analysis. In this paper, a joint filter approach is proposed to estimate both the static states (voltage magnitude and phase angle of bus) and the dynamic states (rotor angle, rotor speed, and transient voltages of generator) utilizing the measurements of phasor measurement unit. First, due to the complexity and dynamics of power system, the power system model is usually not known accurately. The uncertainties include both model uncertainty and noise statistics uncertainty. Therefore, H∞ filter is applied to reliable estimation of the static states. As a link between the static state estimation and the dynamic state estimation, the estimated static states are fed to dynamic state estimation. Next, to deal with the nonlinearity of synchronous generator model, an extended Kalman particle filter is employed to accurately estimate the dynamic states. The numerical tests on IEEE 30-bus system show that the proposed method is into both the static state estimation and the dynamic state estimation.
- 7 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 68 (1) , 87-94
A Joint Filter Approach for Reliable Power System State Estimation
Accurate power system state estimation facilitates power system control, optimization, and security analysis. In this paper, a joint filter approach is proposed to estimate both the static states (voltage magnitude and phase angle of bus) and the dynamic states (rotor angle, rotor speed, and transient voltages of generator) utilizing the measurements of phasor measurement unit. First, due to the complexity and dynamics of power system, the power system model is usually not known accurately. The uncertainties include both model uncertainty and noise statistics uncertainty. Therefore, H∞ filter is applied to reliable estimation of the static states. As a link between the static state estimation and the dynamic state estimation, the estimated static states are fed to dynamic state estimation. Next, to deal with the nonlinearity of synchronous generator model, an extended Kalman particle filter is employed to accurately estimate the dynamic states. The numerical tests on IEEE 30-bus system show that the proposed method is into both the static state estimation and the dynamic state estimation.