Hong Zhu
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
Hong Zhu has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
- Meiyi Xie
Huazhong University of Science and Technology
12 shared publications - Zhiqiang Zhang
Chongqing University of Posts and Telecommunications
5 shared publications - Yong Wang
Chongqing University of Posts and Telecommunications
4 shared publications - Bo Yang
School of Information Science and Engineering, Hunan Institute of Science and Technology, Yueyang, Hunan, China
3 shared publications Ji Cao
3 shared publications- Jian-Shu Lou
School of Pharmacy, Hangzhou Normal University, Hangzhou, Zhejiang 311121, China
3 shared publications
Latest Publications
- 21 January 2026
- Published by Elsevier in Neural Networks
- Vol. 199, 108630
Lifelong knowledge graph embedding via diffusion model
- 21 January 2026
- Published by Elsevier in Neural Networks
- Vol. 199, 108630
Lifelong knowledge graph embedding via diffusion model
- 14 November 2025
- Published by Springer Nature in World Wide Web
- Vol. 28 (6) , 74
An efficient knowledge graph storage framework with adaptive query processing and online updating
- 14 November 2025
- Published by Springer Nature in World Wide Web
- Vol. 28 (6) , 74
An efficient knowledge graph storage framework with adaptive query processing and online updating
- 5 June 2025
- Published by Elsevier in Neurocomputing
- Vol. 647
Multi-level aggregation based federated few-shot learning
- 5 June 2025
- Published by Elsevier in Neurocomputing
- Vol. 647
Multi-level aggregation based federated few-shot learning
- 17 January 2025
- Published by Elsevier in Neural Networks
- Vol. 185, 107131
DGMSCL: A dynamic graph mixed supervised contrastive learning approach for class imbalanced multivariate time series classification
- 17 January 2025
- Published by Elsevier in Neural Networks
- Vol. 185, 107131
DGMSCL: A dynamic graph mixed supervised contrastive learning approach for class imbalanced multivariate time series classification
- 10 December 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 55 (2) , 1-18
Multivariate time series classification based on spatial-temporal attention dynamic graph neural network
- 10 December 2024
- Published by Springer Nature in Applied Intelligence
- Vol. 55 (2) , 1-18
Multivariate time series classification based on spatial-temporal attention dynamic graph neural network
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 72 (2) , 2056-2067
Multimodal Chaotification Model With Hardware Implementation
Designing chaotic maps with desired dynamic properties is an attractive but challenging task. Although a lot of efforts have been devoted, many existing chaotic maps still suffer from some problems such as the insufficient chaotic behavior, discontinuous chaotic range, and nonuniform sequence distribution, which may bring negative effects to the industrial applications based on them. To alleviate these problems, this article proposes a multimodal chaotification model (MCM), and it can generate many new 1-D enhanced chaotic maps by performing the specially designed geometric and linear operations on existing 1-D chaotic maps. Based on the Lyapunov exponent and Frobenius–Perron operator, MCM is analyzed, and a series of theorems and corollaries obtained theoretically proves its effectiveness. As examples, two new 1-D chaotic maps are generated by our MCM, and the numerical evaluations demonstrate their outstanding properties. At the same time, the experimental results of field-programmable gate array also show that they have good implementability on hardware devices. To show the potential of MCM in industrial applications, the illustrative chaotic maps are applied to pseudorandom number generator and image encryption, and our designed schemes achieve superior performance indicators.
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 72 (2) , 2056-2067
Multimodal Chaotification Model With Hardware Implementation
Designing chaotic maps with desired dynamic properties is an attractive but challenging task. Although a lot of efforts have been devoted, many existing chaotic maps still suffer from some problems such as the insufficient chaotic behavior, discontinuous chaotic range, and nonuniform sequence distribution, which may bring negative effects to the industrial applications based on them. To alleviate these problems, this article proposes a multimodal chaotification model (MCM), and it can generate many new 1-D enhanced chaotic maps by performing the specially designed geometric and linear operations on existing 1-D chaotic maps. Based on the Lyapunov exponent and Frobenius–Perron operator, MCM is analyzed, and a series of theorems and corollaries obtained theoretically proves its effectiveness. As examples, two new 1-D chaotic maps are generated by our MCM, and the numerical evaluations demonstrate their outstanding properties. At the same time, the experimental results of field-programmable gate array also show that they have good implementability on hardware devices. To show the potential of MCM in industrial applications, the illustrative chaotic maps are applied to pseudorandom number generator and image encryption, and our designed schemes achieve superior performance indicators.
- 11 January 2024
- Published by Springer Nature
- p. 446-466
Item-Oriented Personalized LDP for Discrete Distribution Estimation
- 11 January 2024
- Published by Springer Nature
- p. 446-466
Item-Oriented Personalized LDP for Discrete Distribution Estimation
- 8 November 2023
- Published by Elsevier in Information Sciences
- Vol. 654
Differential privacy may have a potential optimization effect on some swarm intelligence algorithms besides privacy-preserving
- 8 November 2023
- Published by Elsevier in Information Sciences
- Vol. 654
Differential privacy may have a potential optimization effect on some swarm intelligence algorithms besides privacy-preserving
- Open Access
- 8 June 2023
- Published by MDPI AG in Preprints.org
Item-Oriented Personalized LDP for Discrete Distribution Estimation
Discrete distribution estimation is a fundamental statistical tool, which is widely used to perform data analysis tasks in various applications involving sensitive personal information. Due to privacy concerns, individuals may not always provide their raw information, which leads to unpredictable biases in the final results of estimated distribution. Local Differential Privacy (LDP) is an advanced technique for privacy protection of discrete distribution estimation. Currently, typical LDP mechanisms provide same protection for all items in the domain, which imposes unnecessary perturbation on less sensitive items and thus degrades the utility of final results. Although, several recent works try to alleviate this problem, the utility can be further improved. In this paper, we propose a novel notion called Item-Oriented Personalized LDP (IPLDP), which independently perturbs different items with different privacy budgets to achieve personalized privacy protection. Furthermore, to satisfy IPLDP, we propose the Item-Oriented Personalized Randomized Response (IPRR) based on the observation that the sensitivity of data shows an inverse relationship with the population size of respective individuals. Theoretical analysis and experimental results demonstrate that our method can provide fine-grained privacy protection and improve data utility simultaneously.
- 8 June 2023
- Published by MDPI AG in Preprints.org
Open AccessItem-Oriented Personalized LDP for Discrete Distribution Estimation
Discrete distribution estimation is a fundamental statistical tool, which is widely used to perform data analysis tasks in various applications involving sensitive personal information. Due to privacy concerns, individuals may not always provide their raw information, which leads to unpredictable biases in the final results of estimated distribution. Local Differential Privacy (LDP) is an advanced technique for privacy protection of discrete distribution estimation. Currently, typical LDP mechanisms provide same protection for all items in the domain, which imposes unnecessary perturbation on less sensitive items and thus degrades the utility of final results. Although, several recent works try to alleviate this problem, the utility can be further improved. In this paper, we propose a novel notion called Item-Oriented Personalized LDP (IPLDP), which independently perturbs different items with different privacy budgets to achieve personalized privacy protection. Furthermore, to satisfy IPLDP, we propose the Item-Oriented Personalized Randomized Response (IPRR) based on the observation that the sensitivity of data shows an inverse relationship with the population size of respective individuals. Theoretical analysis and experimental results demonstrate that our method can provide fine-grained privacy protection and improve data utility simultaneously.