Hua Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Hua Xu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Chongyang Yang
School of Artificial Intelligence and Computer Science, Jiangnan University, China
1 shared publication - Jinfeng Yang
Department of Artificial Intelligence and Computer Science, Jiangnan University, WuXi, China
1 shared publication - Rui Li
Jiangnan University, China
1 shared publication - YiFan Gu
Jiangnan University
1 shared publication - Zidan Zhang
School of Artificial Intelligence and Computer Science, Jiangnan University, China
1 shared publication
Latest Publications
- Open Access
- 26 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 28-34
Density Peak Clustering Algorithm Based on Natural Neighbor and Diffusion Assignment
Density Peaks Clustering (DPC) algorithm identifies cluster centers based on local density and relative distance, disregarding the effect of the sample's environment on its density. This makes it challenging to identify cluster centers in low-density regions. Moreover, the single-step allocation strategy of DPC algorithm lacks fault tolerance. If a sample point is incorrectly allocated, it can trigger a cascade of subsequent allocation error. To solve the aforementioned issue, a density peak clustering algorithm based on natural neighbors and diffusion allocation (NNDA-DPC) is proposed. Firstly, the concept of natural neighbors is introduced, and the local density of sample points is redefined by considering the density of sample points and the environment in which the sample points are located. This is done to reduce the influence of the sparsity of class clusters on the selection of class cluster centers. The clustering centers are automatically determined by computing the decision value. Secondly, the natural neighbor points of the selected clustering centers are assigned to the corresponding clustering centers to form the initial microclusters. Finally, the initial microclusters are formed based on the remaining sample points and the initial clusters. The final clustering result is obtained by assigning the sample points based on the correlation between the remaining sample points and the initial clusters, which helps prevent the cascading effect of assignment errors. We compared the NNDA-DPC algorithm with DPC and its enhanced versions on artificial and UCI datasets. The experimental results demonstrate that the NNDA-DPC algorithm can efficiently cluster the datasets.
- 26 July 2024
- Published by Association for Computing Machinery (ACM)
- p. 28-34
Open AccessDensity Peak Clustering Algorithm Based on Natural Neighbor and Diffusion Assignment
Density Peaks Clustering (DPC) algorithm identifies cluster centers based on local density and relative distance, disregarding the effect of the sample's environment on its density. This makes it challenging to identify cluster centers in low-density regions. Moreover, the single-step allocation strategy of DPC algorithm lacks fault tolerance. If a sample point is incorrectly allocated, it can trigger a cascade of subsequent allocation error. To solve the aforementioned issue, a density peak clustering algorithm based on natural neighbors and diffusion allocation (NNDA-DPC) is proposed. Firstly, the concept of natural neighbors is introduced, and the local density of sample points is redefined by considering the density of sample points and the environment in which the sample points are located. This is done to reduce the influence of the sparsity of class clusters on the selection of class cluster centers. The clustering centers are automatically determined by computing the decision value. Secondly, the natural neighbor points of the selected clustering centers are assigned to the corresponding clustering centers to form the initial microclusters. Finally, the initial microclusters are formed based on the remaining sample points and the initial clusters. The final clustering result is obtained by assigning the sample points based on the correlation between the remaining sample points and the initial clusters, which helps prevent the cascading effect of assignment errors. We compared the NNDA-DPC algorithm with DPC and its enhanced versions on artificial and UCI datasets. The experimental results demonstrate that the NNDA-DPC algorithm can efficiently cluster the datasets.
- Open Access
- 26 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 470-477
Memory-based adaptive MOEAD algorithm for multi-objective green fuzzy flexible job shop scheduling problem
This paper introduces an adaptive multi-objective evolutionary algorithm (MBEA) with a memory mechanism to solve the resolution of the fuzzy flexible job shop scheduling problem. The algorithm aims to minimize both the maximum completion time and energy consumption. In contrast to the conventional polling local search approach, the algorithm employs a technique of dynamically choosing local search strategies. This enhances the overall optimization capacity and quality of solutions, while also reducing the search duration. This work presents a memory-based strategy as a solution to the issue commonly observed in traditional MOEA/D, where potential high-quality solutions are often lost during the iteration process. The proposed strategy aims to store solutions with promising properties, thereby enhancing the convergence speed of the algorithm and mitigating the risk of the algorithm getting stuck in local optimization. A comprehensive strategy is proposed that incorporates five initialization rules to ensure the generation of a high-quality initial population. This strategy is further enhanced by implementing an elite archiving strategy to effectively cover the Pareto frontiers. Extensive experimentation was carried out on 23 datasets of varying sizes. The results of the experiments indicate that the enhanced algorithm exhibits notably superior performance compared to the other five widely recognized evolutionary algorithms. This research outcome not only introduces a novel and efficient approach for addressing the fuzzy flexible job shop scheduling issue but also holds significant practical value and provides guidance for scheduling decisions in real-world industrial production scenarios.
- 26 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 470-477
Open AccessMemory-based adaptive MOEAD algorithm for multi-objective green fuzzy flexible job shop scheduling problem
This paper introduces an adaptive multi-objective evolutionary algorithm (MBEA) with a memory mechanism to solve the resolution of the fuzzy flexible job shop scheduling problem. The algorithm aims to minimize both the maximum completion time and energy consumption. In contrast to the conventional polling local search approach, the algorithm employs a technique of dynamically choosing local search strategies. This enhances the overall optimization capacity and quality of solutions, while also reducing the search duration. This work presents a memory-based strategy as a solution to the issue commonly observed in traditional MOEA/D, where potential high-quality solutions are often lost during the iteration process. The proposed strategy aims to store solutions with promising properties, thereby enhancing the convergence speed of the algorithm and mitigating the risk of the algorithm getting stuck in local optimization. A comprehensive strategy is proposed that incorporates five initialization rules to ensure the generation of a high-quality initial population. This strategy is further enhanced by implementing an elite archiving strategy to effectively cover the Pareto frontiers. Extensive experimentation was carried out on 23 datasets of varying sizes. The results of the experiments indicate that the enhanced algorithm exhibits notably superior performance compared to the other five widely recognized evolutionary algorithms. This research outcome not only introduces a novel and efficient approach for addressing the fuzzy flexible job shop scheduling issue but also holds significant practical value and provides guidance for scheduling decisions in real-world industrial production scenarios.