Bo Zhang
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
Bo Zhang has a total of 146 co-authors. Here are the top co-authors by number of shared publications:
Meizi Li
12 shared publications- Ru Yang
Department of Computer Science and Technology, Shanghai Normal University, Shanghai, China
8 shared publications - Maozhen Li
Department of Electronic and Electrical Engineering, Brunel University London, Uxbridge UB8 3PH, UK
7 shared publications - Meizi Li
College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai, China
7 shared publications - Jianguo Pan
College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China
4 shared publications - Ling Ding
Tongji University Shanghai China
4 shared publications
Latest Publications
- 3 October 2026
- Published by Elsevier in Information Sciences
- Vol. 762
HeRaKT: Heterogeneous bidirectional modeling for response-aware knowledge tracing
- 3 October 2026
- Published by Elsevier in Information Sciences
- Vol. 762
HeRaKT: Heterogeneous bidirectional modeling for response-aware knowledge tracing
- 19 August 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
SIGNAL: Synergistic multi-granularity semantic alignment learning for multimodal recommendation
- 19 August 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 333
SIGNAL: Synergistic multi-granularity semantic alignment learning for multimodal recommendation
- Open Access
- 14 July 2026
- Published by Elsevier in SSRN Electronic Journal
Decarbonizing Long-haul Heavy-duty Electric Trucks with Co-optimized Corridor Charging Hubs
The decarbonization of long-haul road freight is a major challenge because high-power charging, grid constraints and corridor-level heterogeneity jointly shape the cost and feasibility of heavy battery electric truck operations. Highway charging hubs can function as en-route charging stops that segment long trunk expressways into feasible charging corridors, but deciding where to build hubs and how large to build them remains difficult when land premiums, grid interconnection costs and time-of-use electricity tariffs vary sharply across space. Here we develop a corridor optimization framework that jointly plans charging-hub location and capacity, grid interconnection, and charging operations. Across 35 selected national trunk expressway freight corridors with one-way distances of 773-4,081 km, we find that optimized corridor planning brings operating costs close to diesel benchmarks, with a mean unit energy cost of about 2.62 CNY km^{-1}, and increases the share of economically competitive corridors from about 17% to 60%, corresponding to 6 to 21 routes. Environmental performance depends strongly on regional grid carbon intensity. Under baseline operations, only 15 of 35 corridors achieve net emissions benefits relative to diesel, whereas optimized planning reduces route-level emissions by about 0.3-50.4% relative to unoptimized HBET operations across all 35 corridors and expands the set of environmentally competitive routes from 15 to 32. Optimized planning also mitigates the payload penalty by reducing required onboard energy storage, with a 60-70% reduction on an ultra-long-haul corridor relative to unoptimized baselines.
- 14 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessDecarbonizing Long-haul Heavy-duty Electric Trucks with Co-optimized Corridor Charging Hubs
The decarbonization of long-haul road freight is a major challenge because high-power charging, grid constraints and corridor-level heterogeneity jointly shape the cost and feasibility of heavy battery electric truck operations. Highway charging hubs can function as en-route charging stops that segment long trunk expressways into feasible charging corridors, but deciding where to build hubs and how large to build them remains difficult when land premiums, grid interconnection costs and time-of-use electricity tariffs vary sharply across space. Here we develop a corridor optimization framework that jointly plans charging-hub location and capacity, grid interconnection, and charging operations. Across 35 selected national trunk expressway freight corridors with one-way distances of 773-4,081 km, we find that optimized corridor planning brings operating costs close to diesel benchmarks, with a mean unit energy cost of about 2.62 CNY km^{-1}, and increases the share of economically competitive corridors from about 17% to 60%, corresponding to 6 to 21 routes. Environmental performance depends strongly on regional grid carbon intensity. Under baseline operations, only 15 of 35 corridors achieve net emissions benefits relative to diesel, whereas optimized planning reduces route-level emissions by about 0.3-50.4% relative to unoptimized HBET operations across all 35 corridors and expands the set of environmentally competitive routes from 15 to 32. Optimized planning also mitigates the payload penalty by reducing required onboard energy storage, with a 60-70% reduction on an ultra-long-haul corridor relative to unoptimized baselines.
- 22 June 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 331
Long- and Short-Term Sequential Recommendation based Transformer and Mixture of Experts
- 22 June 2026
- Published by Elsevier in Expert Systems with Applications
- Vol. 331
Long- and Short-Term Sequential Recommendation based Transformer and Mixture of Experts
- 1 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13  (3) , 3288-3305
Uncertainty-Aware Bayesian Graph Convolutional Network With Sequential Transfer Learning: A Framework for Gravitational Source Identification in Social Networks
Social networks exhibit a Gravitational Field phenomenon, where key nodes, called Gravitational Sources, generate influential zones in their neighborhoods. Traditional methods mainly rely on centrality metrics or heuristic approaches that estimate influence. But they fail to capture the dual characteristics of topology and dynamics required for accurately identifying Gravitational Sources. Identifying Gravitational Sources faces three main challenges: 1) how to construct an effective evaluation mechanism to calculate gravitational source scores; 2) how to achieve effective model training in large-scale network environments with label scarcity; and 3) how to quantify the uncertainty of prediction results to enhance identification reliability. To address these challenges, we propose uncertainty-aware Bayesian graph convolutional network with sequential transfer learning (UBGCN-STL). Our approach integrates an evaluation method that combines centrality and influence metrics, employs a sequential transfer learning strategy to leverage knowledge from label-rich small networks for large networks, and incorporates a Bayesian mechanism for uncertainty-aware predictions. Experiments on seven real-world networks (social, protein interaction, and animal networks) show that UBGCN-STL outperforms representative baseline methods based on centrality, network structure, or learning models (LCNN, NDM, IpGCN), achieving superior predictive performance and reliable uncertainty quantification even with limited labels.
- 1 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13  (3) , 3288-3305
Uncertainty-Aware Bayesian Graph Convolutional Network With Sequential Transfer Learning: A Framework for Gravitational Source Identification in Social Networks
Social networks exhibit a Gravitational Field phenomenon, where key nodes, called Gravitational Sources, generate influential zones in their neighborhoods. Traditional methods mainly rely on centrality metrics or heuristic approaches that estimate influence. But they fail to capture the dual characteristics of topology and dynamics required for accurately identifying Gravitational Sources. Identifying Gravitational Sources faces three main challenges: 1) how to construct an effective evaluation mechanism to calculate gravitational source scores; 2) how to achieve effective model training in large-scale network environments with label scarcity; and 3) how to quantify the uncertainty of prediction results to enhance identification reliability. To address these challenges, we propose uncertainty-aware Bayesian graph convolutional network with sequential transfer learning (UBGCN-STL). Our approach integrates an evaluation method that combines centrality and influence metrics, employs a sequential transfer learning strategy to leverage knowledge from label-rich small networks for large networks, and incorporates a Bayesian mechanism for uncertainty-aware predictions. Experiments on seven real-world networks (social, protein interaction, and animal networks) show that UBGCN-STL outperforms representative baseline methods based on centrality, network structure, or learning models (LCNN, NDM, IpGCN), achieving superior predictive performance and reliable uncertainty quantification even with limited labels.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Education
- Vol. 69  (4) , 1-13
MPHINE-KT: Heterogeneous Information Network Embedding With Metapath Knowledge Tracing
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Education
- Vol. 69  (4) , 1-13
MPHINE-KT: Heterogeneous Information Network Embedding With Metapath Knowledge Tracing
- 25 October 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 330
Dual contrastive learning-based hypergraph convolutional network for aspect-based sentiment classification
- 25 October 2025
- Published by Elsevier in Knowledge-Based Systems
- Vol. 330
Dual contrastive learning-based hypergraph convolutional network for aspect-based sentiment classification
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (3) , 3215-3228
MagPrint++ : Continuous User Fingerprinting on Mobile Devices Using Electromagnetic Signals
Understanding the nature of user-device interactions (e.g., who is using the device and what he/she is doing with it) is critical for many applications including time management, user profiles, and privacy protection. However, in scenarios where mobile devices are shared among family members or multiple employees in a company, conventional account-based statistics are not meaningful. This poses an even bigger problem when dealing with sensitive data. Moreover, fingerprint readers and front-facing cameras were not designed to continuously identify users. In this study, we developed MagPrint++, a novel approach to fingerprint users based on unique patterns in the electromagnetic (EM) signals associated with the specific use patterns of users. Initial experiments showed that time-varying EM patterns are unique to individual users. They are also temporally and spatially consistent, which makes them suitable for fingerprinting. MagPrint++ has a number of advantages over existing schemes: i) Non-intrusive fingerprinting, ii) implementation both on COTS mobile phones and a small and easy-to-deploy device, and iii) high accuracy thanks to the proposed classification algorithm. In experiments involving 30 users, MagPrint++ achieves 94.3% accuracy in classifying users from these traces, which represents a 10.9% improvement over the state-of-the-art classification method.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (3) , 3215-3228
MagPrint++ : Continuous User Fingerprinting on Mobile Devices Using Electromagnetic Signals
Understanding the nature of user-device interactions (e.g., who is using the device and what he/she is doing with it) is critical for many applications including time management, user profiles, and privacy protection. However, in scenarios where mobile devices are shared among family members or multiple employees in a company, conventional account-based statistics are not meaningful. This poses an even bigger problem when dealing with sensitive data. Moreover, fingerprint readers and front-facing cameras were not designed to continuously identify users. In this study, we developed MagPrint++, a novel approach to fingerprint users based on unique patterns in the electromagnetic (EM) signals associated with the specific use patterns of users. Initial experiments showed that time-varying EM patterns are unique to individual users. They are also temporally and spatially consistent, which makes them suitable for fingerprinting. MagPrint++ has a number of advantages over existing schemes: i) Non-intrusive fingerprinting, ii) implementation both on COTS mobile phones and a small and easy-to-deploy device, and iii) high accuracy thanks to the proposed classification algorithm. In experiments involving 30 users, MagPrint++ achieves 94.3% accuracy in classifying users from these traces, which represents a 10.9% improvement over the state-of-the-art classification method.
- 4 October 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 162
Enhanced air pollution spatiotemporal forecast model using frequency domain convolution and attention mechanism
- 4 October 2025
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 162
Enhanced air pollution spatiotemporal forecast model using frequency domain convolution and attention mechanism
- 12 July 2025
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67  (11) , 9779-9820
A unified review of aspect sentiment triplet extraction methods in aspect-based sentiment analysis
- 12 July 2025
- Published by Springer Nature in Knowledge and Information Systems
- Vol. 67  (11) , 9779-9820
A unified review of aspect sentiment triplet extraction methods in aspect-based sentiment analysis