Hui Du
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Hui Du has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Guanghao Jin
School of Artificial Intelligence, Beijing Polytechnic, China
5 shared publications - Qingzeng Song
Tiangong University, Tianjin, China
4 shared publications - Yuqing Wang
School of Computer Science and Technology, Tiangong University, Tianjin, China
4 shared publications - Jieying Wang
School of Computer Science and Technology, Tiangong University, China
3 shared publications - Lei Ma
School of Artificial Intelligence, Beijing Polytechnic, China
2 shared publications - Hengguang Li
Department of Mathematics, Wayne State University, Detroit, USA
1 shared publication
Latest Publications
- Open Access
- 16 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 14-18
Anomaly Sample Detection for Deep Learning based Teaching Data Collection
This paper introduces an anomaly sample detection method for the high quality collection of teaching data. Our method uses a deep learning method to detect the anomaly samples of the collected teaching samples. Then, we construct a training set for the deep learning models by filtering these samples. We use this training set to train the deep learning models for high quality teaching data collection. The experiment results prove the effectiveness of our method both on public datasets and teaching samples.
- 16 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 14-18
Open AccessAnomaly Sample Detection for Deep Learning based Teaching Data Collection
This paper introduces an anomaly sample detection method for the high quality collection of teaching data. Our method uses a deep learning method to detect the anomaly samples of the collected teaching samples. Then, we construct a training set for the deep learning models by filtering these samples. We use this training set to train the deep learning models for high quality teaching data collection. The experiment results prove the effectiveness of our method both on public datasets and teaching samples.
- Open Access
- 1 January 2025
- Published by Wiley in Complexity
- Vol. 2025 (1)
Text Classification of Multiple Datasets Based on Multidimensional Information
Multiple datasets enable a deep learning model to achieve a wide range of classifications, while the diversity of datasets reduces classification accuracy. To solve this problem, a method based on multidimensional information is proposed. The first dimension is the outputs of multiple models on different datasets. Through this information, we can predict the dataset that may contain the testing samples. The second one is the outputs of multiple models on the same dataset, through which the labels of testing samples can be classified. The third one is the distribution of labels on the same testing sample, which further increases accuracy. Experimental results show that our method achieves the best performance compared to the existing methods while ensuring good scalability.
- 1 January 2025
- Published by Wiley in Complexity
- Vol. 2025 (1)
Open AccessText Classification of Multiple Datasets Based on Multidimensional Information
Multiple datasets enable a deep learning model to achieve a wide range of classifications, while the diversity of datasets reduces classification accuracy. To solve this problem, a method based on multidimensional information is proposed. The first dimension is the outputs of multiple models on different datasets. Through this information, we can predict the dataset that may contain the testing samples. The second one is the outputs of multiple models on the same dataset, through which the labels of testing samples can be classified. The third one is the distribution of labels on the same testing sample, which further increases accuracy. Experimental results show that our method achieves the best performance compared to the existing methods while ensuring good scalability.
- Open Access
- 26 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 220-226
Multi-dataset Pose Estimation based on the Fusion of Multiple Models
To increase the accuracy of pose estimation on multiple datasets, this paper proposes a new pose estimation method based on the fusion of multiple models. This method consists of two main steps: firstly, a dataset classification by YoloV5 algorithm is performed that is to classify the dataset that may contain the testing samples; secondly, the key point detection is performed by using a deep learning model-based pose estimation that is trained on the predicted dataset. We evaluated the methods on three publicly datasets: Leeds Sports Pose (LSP) dataset, CrowdPose dataset and AnimalPose one. As the experimental results show, our method can achieve the scalability on multiple datasets while ensuring high accuracy.
- 26 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 220-226
Open AccessMulti-dataset Pose Estimation based on the Fusion of Multiple Models
To increase the accuracy of pose estimation on multiple datasets, this paper proposes a new pose estimation method based on the fusion of multiple models. This method consists of two main steps: firstly, a dataset classification by YoloV5 algorithm is performed that is to classify the dataset that may contain the testing samples; secondly, the key point detection is performed by using a deep learning model-based pose estimation that is trained on the predicted dataset. We evaluated the methods on three publicly datasets: Leeds Sports Pose (LSP) dataset, CrowdPose dataset and AnimalPose one. As the experimental results show, our method can achieve the scalability on multiple datasets while ensuring high accuracy.
- Open Access
- 26 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 36-41
A System to Increase the Accuracy of Deep Learning based Classification by the Distribution of Labels
In some deep learning-based applications, we can use the distribution of labels to increase the classification accuracy like the distribution of the labels in the corresponding application. This kind distribution can be obtained by the experience of human being while it is generally imprecise. Using validation set can be a solution to compute precise distribution of labels while this is affected by the application change. In this paper, we propose a novel system, which efficiently utilizes the experience of human being and validation set to improve the performance of classification models. Firstly, our method selects and trains an existing deep learning model on the training set. Then, these models can output the probability of labels on each testing sample. Secondly, our system utilizes the user input and the validation set, which is to get the precise distribution of labels in an application. Finally, the computed distribution of labels is selectively combined with the probability of labels on testing samples to get the final classification results. We selected some existing deep learning models on three public datasets to evaluate the models. As the evaluation results show, our system can improve the performance of classification tasks, which is obviously higher than state-of-art methods.
- 26 January 2024
- Published by Association for Computing Machinery (ACM)
- p. 36-41
Open AccessA System to Increase the Accuracy of Deep Learning based Classification by the Distribution of Labels
In some deep learning-based applications, we can use the distribution of labels to increase the classification accuracy like the distribution of the labels in the corresponding application. This kind distribution can be obtained by the experience of human being while it is generally imprecise. Using validation set can be a solution to compute precise distribution of labels while this is affected by the application change. In this paper, we propose a novel system, which efficiently utilizes the experience of human being and validation set to improve the performance of classification models. Firstly, our method selects and trains an existing deep learning model on the training set. Then, these models can output the probability of labels on each testing sample. Secondly, our system utilizes the user input and the validation set, which is to get the precise distribution of labels in an application. Finally, the computed distribution of labels is selectively combined with the probability of labels on testing samples to get the final classification results. We selected some existing deep learning models on three public datasets to evaluate the models. As the evaluation results show, our system can improve the performance of classification tasks, which is obviously higher than state-of-art methods.
- Open Access
- 22 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 155-159
Multi-level Teaching Text Classification based on the Fusion of Deep Learning Models
- 22 December 2023
- Published by Association for Computing Machinery (ACM)
- p. 155-159
Open AccessMulti-level Teaching Text Classification based on the Fusion of Deep Learning Models