A5100458629- Advanced Graph Neural Networks
- Recommender Systems and Techniques
- Smart Grid Energy Management
- Advanced Data Storage Technologies
- Evolutionary Algorithms and Applications
- Advanced Text Analysis Techniques
- Advanced Computational Techniques and Applications
- Multimodal Machine Learning Applications
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Digital Marketing and Social Media
- Authorship Attribution and Profiling
- Publishing and Scholarly Communication
- Biochemical and Structural Characterization
- Hearing, Cochlea, Tinnitus, Genetics
- Machine Learning and Data Classification
- Chaos-based Image/Signal Encryption
- Neuroscience and Neural Engineering
- 3D Printing in Biomedical Research
- Wireless Communication Security Techniques
- Machine Fault Diagnosis Techniques
- Software System Performance and Reliability
- Micro and Nano Robotics
- Topic Modeling
- Web Data Mining and Analysis
Beihang University
2018-2023
Hebei University of Technology
2022
Shijiazhuang Tiedao University
2022
Minzu University of China
2021
Nanjing University
2018
City University of Hong Kong
2018
North China University of Science and Technology
2008
Graph Neural Networks (GNNs) have been widely used for the representation learning of various structured graph data, typically through message passing among nodes by aggregating their neighborhood information via different operations. While promising, most existing GNNs oversimplify complexity and diversity edges in thus are inefficient to cope with ubiquitous heterogeneous graphs, which form multi-relational representations. In this article, we propose RioGNN , a novel Reinforced,...
Massive open online courses (MOOCs) , which offer access and widespread interactive participation through the internet, are quickly becoming preferred method for remote learning. Several MOOC platforms service of course recommendation to users, improve learning experience users. Despite usefulness this service, we consider that recommending users directly may neglect their varying degrees expertise. To mitigate gap, examine an interesting problem concept in paper, can be viewed as knowledge...
DBSCAN is widely used in many scientific and engineering fields because of its simplicity practicality. However, due to high sensitivity parameters, the accuracy clustering result depends heavily on practical experience. In this paper, we first propose a novel Deep Reinforcement Learning guided automatic parameters search framework, namely DRL-DBSCAN. The framework models process adjusting parameter direction by perceiving environment as Markov decision process, which aims find best without...
Gearbox with complex structure is one of the most fragile components wind turbines. Fault diagnosis gearbox crucial to reduce unexpected downtime and economic losses. This paper proposes an intelligent fault method based on Long Short-term Memory (LSTM) networks. Firstly, multi- accelerometers vibration signals are divided into data segments. Then common time domain features extracted from these After that, fed LSTM networks for pattern classification. The proposed has no requirement...
<abstract><p>Cognitive green computing (CGC) is widely used in the Internet of Things (IoT) for smart city. As power system city, grid has benefited from CGC, which can achieve dynamic regulation electric energy and resource integration optimization. However, it still challenging improving identification accuracy performance load model grid. In this paper, we present a novel algorithm framework based on reinforcement learning (RL) to improve non-invasive monitoring (NILMI)....
Social networks are quickly becoming the primary medium for discussing what is happening around real-world events. However, it still a challenge to detect events on social media due its real-time nature, scale and amount of unstructured data generated. In this paper, we present novel system detecting surrounding Our proposed framework consists four main components, including text filtering, representation, deep clustering, event merging. After filtering non-event messages, use entities words...
Abstract The rapid development of the Internet has brought convenience to people and also produced problem “information overload”. In view traditional collaborative filtering algorithm facing some bottlenecks be solved, this study proposes a that combines similarity trust. First all, in large deviation calculation prediction user ratings, an optimized Pearson correlation coefficient method; secondly, trust relationship is established based on user’s rating common project, between users who...