A5103325773- Recommender Systems and Techniques
- Cloud Computing and Resource Management
- Software System Performance and Reliability
- Software Engineering Research
- Advanced Graph Neural Networks
- Coastal Management and Development
- Syntax, Semantics, Linguistic Variation
- Advanced Adaptive Filtering Techniques
- Power Line Communications and Noise
- Distributed and Parallel Computing Systems
- Software Testing and Debugging Techniques
- Big Data and Business Intelligence
- Advanced Technologies in Various Fields
- Marine and Coastal Research
- Advanced Text Analysis Techniques
- Complex Network Analysis Techniques
- Big Data Technologies and Applications
- Remote Sensing and Land Use
- Data Stream Mining Techniques
- Software Reliability and Analysis Research
- Language, Metaphor, and Cognition
- Online Learning and Analytics
- Natural Language Processing Techniques
- Text and Document Classification Technologies
- Radio Wave Propagation Studies
Qingdao University of Science and Technology
2022-2025
Shanghai Dianji University
2024
Zhejiang Ocean University
2019
Zhejiang University
2019
Southwest University of Political Science & Law
2017
Abstract In computing application scenarios with large volumes of data, time-efficient data warehouses are the primary choice for most businesses. The metadata module will be designed MySQL as an intermediate node information exchange among modules in efficient warehouse this paper. first and second-layer scheduling algorithms utilized to monitor progress queries updates system real-time, realize intelligent setting dynamic priorities processing tasks. Subsequently, execution is built based...
Multi-behavior recommendation systems enhance effectiveness by leveraging auxiliary behaviors (such as page views and favorites) to address the limitations of traditional models that depend solely on sparse target like purchases. Existing approaches multi-behavior recommendations typically follow one two strategies: some derive initial node representations from individual behavior subgraphs before integrating them for a comprehensive profile, while others interpret data heterogeneous graph,...
Studies in the past mainly focus on garden path phenomenon from perspective of cognitive linguistics and psycholinguistics. The adequacies their explanations are different, but these theories imperfect. This paper discusses generative grammar. author holds that θ-attachment principle can analyze reason partial ambiguity effectively provides evidence for derivation by phase under framework MP People clarify structure succinctly enhance understanding language mechanism through analysis them...
Abstract Generally, open source software (OSS) has a longer bug‐fixing time. If the time can be predicted accurately as early possible, it will beneficial to efficiency of bug fixing. Traditional prediction models are usually based on static features report. It is difficult go into service due inappropriate feature extraction data and low accuracy models. The HMM model predict according earlier fixing activities. However, this method temporal sequence selection results in large number...
Disentangled collaborative filtering can explicitly generate embeddings based on users' interests and help improve the interpretability robustness of recommendations. However, existing disentangled graph methods rely solely direct interaction constraints between nodes to learn node embeddings, which cannot represent higher-order node-type differences, resulting in suboptimal representations negatively affecting recommendation performance. To address this problem, we propose a Multi-order...