Zhiwei Ni
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhiwei Ni has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Feifei Jin
School of Business, Anhui University, Hefei, Anhui 230601, P. R. China
2 shared publications - Liyang Tang
Hefei University of Technology
2 shared publications - Ping Li
School of Management, Hefei University of Technology, Hefei 230009, Anhui, China
2 shared publications - Qian Chen
School of Artificial and Big Data, Hefei University, Anhui 230601, China
2 shared publications - Wentao Liu
School of Management, Hefei University of Technology, Hefei 230009, China
2 shared publications - Wenying Wu
School of Management, Hefei University of Technology, Hefei 230009, Anhui, China
2 shared publications
Latest Publications
- Open Access
- 25 March 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (7) , 3556
An Unmanned Delivery Vehicle Path-Planning Method Based on Point-Graph Joint Embedding and Dual Decoders
The path-planning of unmanned delivery vehicles (UDVs) has garnered significant interest due to their extensive use in contactless delivery during severe epidemics and automated delivery of parcels in diverse scenarios. However, previous studies have focused on achieving the shortest path or time based on the comprehensive cost consumption in the transportation process and ignored the impact of different customers’ different delivery time requirements in the actual interactive system. Hence, a path-planning model is presented to tackle the routing dilemma of UDVs in logistics. This new dilemma, called the unmanned delivery vehicle routing problem (UDVRP), considers the comprehensive transportation cost consumption of distribution vehicles and the customer satisfaction of each distribution point. Customer satisfaction is defined based on the delivery time requirements of different customers. This novel deep neural network model incorporates an attention mechanism and applies a method called point-graph joint embedding and dual decoders (PGDD) to solve the problem. The network’s architecture, consisting of an encoder and two decoders, directly determines the path for unmanned delivery vehicles. In addition, the model is trained offline using a deep reinforcement-learning strategy in combination with pseudo-label learning. In this scenario, the output of one decoder serves as the label for another, overseeing its learning process to choose the most effective path. Experimental results demonstrate that PGDD reduces total costs by 8.73% on average compared to state-of-the-art algorithms in 100-node scenarios, with performance gains reaching 12.5% for larger-scale problems (400 nodes), validating its superiority in complex path-planning. Additionally, PGDD improves customer satisfaction by 15.2% and achieves a response time below 90ms in real-world deployment tests. The experimental results demonstrate that the proposed method is superior to several state-of-the-art algorithms in solving the path-planning problem of unmanned distribution vehicles.
- 25 March 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (7) , 3556
Open AccessAn Unmanned Delivery Vehicle Path-Planning Method Based on Point-Graph Joint Embedding and Dual Decoders
The path-planning of unmanned delivery vehicles (UDVs) has garnered significant interest due to their extensive use in contactless delivery during severe epidemics and automated delivery of parcels in diverse scenarios. However, previous studies have focused on achieving the shortest path or time based on the comprehensive cost consumption in the transportation process and ignored the impact of different customers’ different delivery time requirements in the actual interactive system. Hence, a path-planning model is presented to tackle the routing dilemma of UDVs in logistics. This new dilemma, called the unmanned delivery vehicle routing problem (UDVRP), considers the comprehensive transportation cost consumption of distribution vehicles and the customer satisfaction of each distribution point. Customer satisfaction is defined based on the delivery time requirements of different customers. This novel deep neural network model incorporates an attention mechanism and applies a method called point-graph joint embedding and dual decoders (PGDD) to solve the problem. The network’s architecture, consisting of an encoder and two decoders, directly determines the path for unmanned delivery vehicles. In addition, the model is trained offline using a deep reinforcement-learning strategy in combination with pseudo-label learning. In this scenario, the output of one decoder serves as the label for another, overseeing its learning process to choose the most effective path. Experimental results demonstrate that PGDD reduces total costs by 8.73% on average compared to state-of-the-art algorithms in 100-node scenarios, with performance gains reaching 12.5% for larger-scale problems (400 nodes), validating its superiority in complex path-planning. Additionally, PGDD improves customer satisfaction by 15.2% and achieves a response time below 90ms in real-world deployment tests. The experimental results demonstrate that the proposed method is superior to several state-of-the-art algorithms in solving the path-planning problem of unmanned distribution vehicles.
- Open Access
- 22 August 2024
- Published by MDPI AG in Electronics
- Vol. 13 (16) , 3346
Dynamic Edge-Based High-Dimensional Data Aggregation with Differential Privacy
Edge computing enables efficient data aggregation for services like data sharing and analysis in distributed IoT applications. However, uploading dynamic high-dimensional data to an edge server for efficient aggregation is challenging. Additionally, there is the significant risk of privacy leakage associated with direct such data uploading. Therefore, we propose an edge-based differential privacy data aggregation method leveraging progressive UMAP with a dynamic time window based on LSTM (EDP-PUDL). Firstly, a model of the dynamic time window based on a long short-term memory (LSTM) network was developed to divide dynamic data. Then, progressive uniform manifold approximation and projection (UMAP) with differential privacy was performed to reduce the dimension of the window data while preserving privacy. The privacy budget was determined by the data volume and the attribute’s Shapley value, adding DP noise. Finally, the privacy analysis and experimental comparisons demonstrated that EDP-PUDL ensures user privacy while achieving superior aggregation efficiency and availability compared to other algorithms used for dynamic high-dimensional data aggregation.
- 22 August 2024
- Published by MDPI AG in Electronics
- Vol. 13 (16) , 3346
Open AccessDynamic Edge-Based High-Dimensional Data Aggregation with Differential Privacy
Edge computing enables efficient data aggregation for services like data sharing and analysis in distributed IoT applications. However, uploading dynamic high-dimensional data to an edge server for efficient aggregation is challenging. Additionally, there is the significant risk of privacy leakage associated with direct such data uploading. Therefore, we propose an edge-based differential privacy data aggregation method leveraging progressive UMAP with a dynamic time window based on LSTM (EDP-PUDL). Firstly, a model of the dynamic time window based on a long short-term memory (LSTM) network was developed to divide dynamic data. Then, progressive uniform manifold approximation and projection (UMAP) with differential privacy was performed to reduce the dimension of the window data while preserving privacy. The privacy budget was determined by the data volume and the attribute’s Shapley value, adding DP noise. Finally, the privacy analysis and experimental comparisons demonstrated that EDP-PUDL ensures user privacy while achieving superior aggregation efficiency and availability compared to other algorithms used for dynamic high-dimensional data aggregation.
- Open Access
- 5 January 2021
- Published by MDPI AG in Mathematics
- Vol. 9 (1) , 107
Investment Selection Based on Bonferroni Mean under Generalized Probabilistic Hesitant Fuzzy Environments
In investment selection problems, the existence of contingency and uncertainty may result in the loss of attribute information. Then, how to make proper investment decision-making will be a tricky proposition. In this work, a multiattribute group decision making (MAGDM) method based on the generalized probabilistic hesitant fuzzy Bonferroni mean (GPHFBM) operator is constructed, which enables decision-makers to select the proper parameters in decision-making process. Firstly, the GPHFBM operator is proposed by combining the Bonferroni mean operator and Archimedean norm. Secondly, five excellent properties of the GPHFBM operator are discussed in detail. In view of applications, we further develop some special aggregation operators for GPHFBM with the various values of parameters b, d and additive operators g(t). Finally, we propose a probabilistic hesitant fuzzy MAGDM method based on the GPHFBM operator to analyze the aggregated information. A case study of the investment of social insurance funds is given to depict the validity and reasonability of the proposed method. Ultimately, the company is selected as the investment company with the best comprehensive indicator.- 5 January 2021
- Published by MDPI AG in Mathematics
- Vol. 9 (1) , 107
Open AccessInvestment Selection Based on Bonferroni Mean under Generalized Probabilistic Hesitant Fuzzy Environments
In investment selection problems, the existence of contingency and uncertainty may result in the loss of attribute information. Then, how to make proper investment decision-making will be a tricky proposition. In this work, a multiattribute group decision making (MAGDM) method based on the generalized probabilistic hesitant fuzzy Bonferroni mean (GPHFBM) operator is constructed, which enables decision-makers to select the proper parameters in decision-making process. Firstly, the GPHFBM operator is proposed by combining the Bonferroni mean operator and Archimedean norm. Secondly, five excellent properties of the GPHFBM operator are discussed in detail. In view of applications, we further develop some special aggregation operators for GPHFBM with the various values of parameters b, d and additive operators g(t). Finally, we propose a probabilistic hesitant fuzzy MAGDM method based on the GPHFBM operator to analyze the aggregated information. A case study of the investment of social insurance funds is given to depict the validity and reasonability of the proposed method. Ultimately, the company is selected as the investment company with the best comprehensive indicator. - Open Access
Optimal Transport with Dimensionality Reduction for Domain Adaptation
Domain adaptation manages to learn a robust classifier for target domain, using the source domain, but they often follow different distributions. To bridge distribution shift between the two domains, most of previous works aim to align their feature distributions through feature transformation, of which optimal transport for domain adaptation has attract researchers’ interest, as it can exploit the local information of the two domains in the process of mapping the source instances to the target ones by minimizing Wasserstein distance between their feature distributions. However, it may weaken the feature discriminability of source domain, thus degrade domain adaptation performance. To address this problem, this paper proposes a two-stage feature-based adaptation approach, referred to as optimal transport with dimensionality reduction (OTDR). In the first stage, we apply the dimensionality reduction with intradomain variant maximization but source intraclass compactness minimization, to separate data samples as much as possible and enhance the feature discriminability of the source domain. In the second stage, we leverage optimal transport-based technique to preserve the local information of the two domains. Notably, the desirable properties in the first stage can mitigate the degradation of feature discriminability of the source domain in the second stage. Extensive experiments on several cross-domain image datasets validate that OTDR is superior to its competitors in classification accuracy.
Open AccessOptimal Transport with Dimensionality Reduction for Domain Adaptation
Domain adaptation manages to learn a robust classifier for target domain, using the source domain, but they often follow different distributions. To bridge distribution shift between the two domains, most of previous works aim to align their feature distributions through feature transformation, of which optimal transport for domain adaptation has attract researchers’ interest, as it can exploit the local information of the two domains in the process of mapping the source instances to the target ones by minimizing Wasserstein distance between their feature distributions. However, it may weaken the feature discriminability of source domain, thus degrade domain adaptation performance. To address this problem, this paper proposes a two-stage feature-based adaptation approach, referred to as optimal transport with dimensionality reduction (OTDR). In the first stage, we apply the dimensionality reduction with intradomain variant maximization but source intraclass compactness minimization, to separate data samples as much as possible and enhance the feature discriminability of the source domain. In the second stage, we leverage optimal transport-based technique to preserve the local information of the two domains. Notably, the desirable properties in the first stage can mitigate the degradation of feature discriminability of the source domain in the second stage. Extensive experiments on several cross-domain image datasets validate that OTDR is superior to its competitors in classification accuracy.
- Open Access
- 6 August 2018
- Published by MDPI AG in Algorithms
- Vol. 11 (8) , 120
Probabilistic Interval-Valued Hesitant Fuzzy Information Aggregation Operators and Their Application to Multi-Attribute Decision Making
Based on the probabilistic interval-valued hesitant fuzzy information aggregation operators, this paper investigates a novel multi-attribute group decision making (MAGDM) model to address the serious loss of information in a hesitant fuzzy information environment. Firstly, the definition of probabilistic interval-valued hesitant fuzzy set will be introduced, and then, using Archimedean norm, some new probabilistic interval-valued hesitant fuzzy operations are defined. Secondly, based on these operations, the generalized probabilistic interval-valued hesitant fuzzy ordered weighted averaging (GPIVHFOWA) operator, and the generalized probabilistic interval-valued hesitant fuzzy ordered weighted geometric (GPIVHFOWG) operator are proposed, and their desirable properties are discussed. We further study their common forms and analyze the relationship among these proposed operators. Finally, a new probabilistic interval-valued hesitant fuzzy MAGDM model is constructed, and the feasibility and effectiveness of the proposed model are verified by using an example of supplier selection.
- 6 August 2018
- Published by MDPI AG in Algorithms
- Vol. 11 (8) , 120
Open AccessProbabilistic Interval-Valued Hesitant Fuzzy Information Aggregation Operators and Their Application to Multi-Attribute Decision Making
Based on the probabilistic interval-valued hesitant fuzzy information aggregation operators, this paper investigates a novel multi-attribute group decision making (MAGDM) model to address the serious loss of information in a hesitant fuzzy information environment. Firstly, the definition of probabilistic interval-valued hesitant fuzzy set will be introduced, and then, using Archimedean norm, some new probabilistic interval-valued hesitant fuzzy operations are defined. Secondly, based on these operations, the generalized probabilistic interval-valued hesitant fuzzy ordered weighted averaging (GPIVHFOWA) operator, and the generalized probabilistic interval-valued hesitant fuzzy ordered weighted geometric (GPIVHFOWG) operator are proposed, and their desirable properties are discussed. We further study their common forms and analyze the relationship among these proposed operators. Finally, a new probabilistic interval-valued hesitant fuzzy MAGDM model is constructed, and the feasibility and effectiveness of the proposed model are verified by using an example of supplier selection.
- 15 January 2015
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 19 (3) , 731-743
Emerging opinion leaders in crowd unfollow crisis: a case study of mobile brands in Twitter
- 15 January 2015
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 19 (3) , 731-743
Emerging opinion leaders in crowd unfollow crisis: a case study of mobile brands in Twitter
- 3 July 2014
- Published by Springer Nature in World Wide Web
- Vol. 18 (4) , 1019-1049
Locating targets through mention in Twitter
- 3 July 2014
- Published by Springer Nature in World Wide Web
- Vol. 18 (4) , 1019-1049
Locating targets through mention in Twitter
- 15 March 2011
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 63 (1) , 256-293
A market-oriented hierarchical scheduling strategy in cloud workflow systems
- 15 March 2011
- Published by Springer Nature in The Journal of Supercomputing
- Vol. 63 (1) , 256-293
A market-oriented hierarchical scheduling strategy in cloud workflow systems
- 9 September 2010
- Published by American Society of Civil Engineers (ASCE)
- p. 3173-3179
Application of Improved Genetic Algorithm in Optimization of Logistics Distribution Routing
The optimization of distribution routing is a key problem of logistics systems optimization, and can be considered as a traveling salesman problem (TSP). In this paper, the optimization of distribution routing is transformed into TSP, and dissimilar order crossover operator is proposed to improve the genetic algorithm (GA) to solve TSP. The improved crossover operator is based on natural representation, which makes the offspring inherit the excellent substrings of its parent and speeds up the convergence of GA. Simultaneously, steady state reproduction without duplicates is adopted to avoid prematurity. It is proved that the improved GA provides prominent performance for both TSP and asymmetric traveling salesman problem (ATSP), which will facilitate the optimization of distribution routing.
- 9 September 2010
- Published by American Society of Civil Engineers (ASCE)
- p. 3173-3179
Application of Improved Genetic Algorithm in Optimization of Logistics Distribution Routing
The optimization of distribution routing is a key problem of logistics systems optimization, and can be considered as a traveling salesman problem (TSP). In this paper, the optimization of distribution routing is transformed into TSP, and dissimilar order crossover operator is proposed to improve the genetic algorithm (GA) to solve TSP. The improved crossover operator is based on natural representation, which makes the offspring inherit the excellent substrings of its parent and speeds up the convergence of GA. Simultaneously, steady state reproduction without duplicates is adopted to avoid prematurity. It is proved that the improved GA provides prominent performance for both TSP and asymmetric traveling salesman problem (ATSP), which will facilitate the optimization of distribution routing.