A5101423488- Complex Network Analysis Techniques
- Infrastructure Maintenance and Monitoring
- Advanced Algorithms and Applications
- Effects of Vibration on Health
- Customer churn and segmentation
- Advanced Computational Techniques and Applications
- Traffic and Road Safety
- Fluid Dynamics Simulations and Interactions
- Oil and Gas Production Techniques
- Advanced Sensor and Control Systems
- Opinion Dynamics and Social Influence
- Advanced Neural Network Applications
- Vehicle Dynamics and Control Systems
- Advanced Graph Neural Networks
- Hydraulic Fracturing and Reservoir Analysis
- Imbalanced Data Classification Techniques
- Fire Detection and Safety Systems
- Drilling and Well Engineering
- Sports Performance and Training
- Soil, Finite Element Methods
- Network Security and Intrusion Detection
- Geographic Information Systems Studies
- Transportation Planning and Optimization
- Innovative Educational Techniques
- 3D Modeling in Geospatial Applications
Changchun University of Science and Technology
2014-2025
Dalian Maritime University
2023-2024
South China University of Technology
2023
Queen's University Belfast
2023
Hefei University of Technology
2017-2018
PRG S&Tech (South Korea)
2014
Changchun University
2014
University of Maine
2011
Lanzhou University of Technology
2007
People's daily lives are situated in both outdoor and indoor space. However, traditional GIS focuses only on Therefore, research providing a unified model of the two spaces making navigation between within them seamless is important. This study lays ontological formal foundations for Four levels ontologies under construction: upper ontology (being constructing by using revising those concepts introduced previous work), domain (structure spaces), task ontology, application ontologies.
Road damage detection is crucial for ensuring road safety and minimizing maintenance costs. However, detecting small damage, managing complex backgrounds, identifying irregular shapes remain significant challenges. To address these issues, we propose YOLO-RD, an advanced framework that integrates innovative modules feature enhancement, multi-scale robustness, detail preservation. Specifically, the Star Operation Module (SOM) improves sensitivity to small-scale Multi-dimensional Auxiliary...
This study explored the relationships between subjective evaluations and objective metrics of vehicle dynamic performance. First, a real test was performed to measure acceleration performance under different conditions, participants’ were investigated. Second, correlation analysis conducted explore each evaluation its corresponding metric as well overall three individual evaluations. Finally, an model related established based on Probabilistic Neural Network (PNN). The results demonstrated...
The infrared and smoke sensors widely applied in fire alarm monitoring the scene of large indoor outdoor warehouse is vulnerable to sorts interference noise, it may out work cause error or omission.Based on design reliable flame recognition algorithm, a practical intelligent system based BP neural network multiple source information was constructed.The uses multi-channel time-division multiplexing technology constitute temperature detector, photoelectric detector CCD into set.The fusion were...
<p>Robust Temporal Link Prediction in Dynamic Complex Networks via Stable Gated with Reinforcement Learning</p>
Temporal link prediction is one of the most important tasks for predicting time-varying links by capturing dynamics within complex networks. However, it suffers from difficulties such as vulnerability to adversarial attacks and inadaptation distinct evolutionary patterns. In this article, we propose a robust temporal architecture via stable gated models with reinforcement learning (SAGE-RL) consisting state encoding network (SEN) self-adaptive policy (SPN). The former utilized capture...
It is reported that the number of online payment users in China has reached 854 million; with emergence community e-commerce platforms, trend integration and social applications increasingly intense. Community not a mature sound comprehensive fewer categories low brand value. To effectively retain fully explore customer value become an important challenge for operators. Given above problems, this paper uses data-driven method to study prediction customers' repurchase behaviour. The main...