A5014864588- Advanced Graph Neural Networks
- Recommender Systems and Techniques
- Welding Techniques and Residual Stresses
- Topic Modeling
- Additive Manufacturing Materials and Processes
- Caching and Content Delivery
- Thermography and Photoacoustic Techniques
- Laser Material Processing Techniques
- Advanced Electron Microscopy Techniques and Applications
- Optical measurement and interference techniques
- Rough Sets and Fuzzy Logic
- Characterization and Applications of Magnetic Nanoparticles
- Image Retrieval and Classification Techniques
- Force Microscopy Techniques and Applications
- Photosynthetic Processes and Mechanisms
- Data Mining Algorithms and Applications
- Magnetic Properties and Applications
- Non-Destructive Testing Techniques
- Advanced Vision and Imaging
- Imbalanced Data Classification Techniques
- 3D Surveying and Cultural Heritage
University College Cork
2023-2025
Xi'an Jiaotong University
2023-2024
Southwest Jiaotong University
2024
Zhengzhou University of Light Industry
2022-2023
In the single-particle Cryo-EM projection image classification, it is a common practice to apply Fourier transform images and extract rotation-invariant features in frequency domain. However, this process involves interpolation, which can reduce accuracy of results. contrast, non-uniform provides more direct accurate computation without need for interpolation process. Leveraging capabilities discrete (NUDFT), we have developed an algorithm classification. To highlight its potential...
With the advent of era rapid information expansion, massive data backlog that exists on Internet has led to a serious overload problem, which makes recommendation systems crucial part human life. In particular, Point-Of-Interest (POI) system been applied many real-life scenarios, such as life services and autonomous driving. Specifically, goal POI is recommend locations match their personalized preferences users. existing methods, people tend pay more attention impact temporal spatial...
In the age of big data, recipe recommendation is great significance. It can recommend recipes in line with user's eating habits massive data. Compared other tasks, influenced by multiple aspects and requires fine-grained learning to obtain entity representations. Therefore, traditional method cannot meet people's requirements. this paper, we propose Collaborative Knowledge Propagation Graph Attention Network for Recipes Recommendation (CGRS). This designs collaborative information...
Macroscopic deformation and microscopic pores are the main defects affecting comprehensive performance of laser powder bed fusion (LPBF) parts. The LPBF process involves highly coupled multiple physical mechanisms over a wide time scale, making it difficult to obtain strongly correlated defect characteristics achieve online monitoring. Thus, this study proposes an acoustic-based macro–micro synchronous monitoring technology. It enables by analysing information progressively from intra-layer...
In the realm of single-particle Cryo-EM's 2D align-ment problem, conventional practice necessitates interpolation as an intermediate step before applying Fourier transform to address in-plane rotation angle. Unfortunately, this interpo-lation process tends introduce errors, making less precise. contrast, Non-uniform offers a more direct and accurate mathematical foundation for analyzing computing angle, eliminating need in related problem-solving tasks. Leveraging capabilities transform, we...
Stereo vision, as a critical part in the field of three-dimensional reconstruction, perceives and measures 3D shape objects by matching corresponding points between left camera right views. The extremely challenging stereo vision is to ensure accuracy areas obtain disparity robustly. However, passive imaging has inherent problems for ill-posed such complicated illumination, weak texture, perspective occlusion adverse environments industrial scenes. active technique designs projects patterns...
During these years, the popularity of Internet, services such as recipe sharing, food reviews and take-out have become increasingly common in life. In scenario using this kind life service, people face problem information overload, so are more inclined to accept system suggestions. Recipe recommendation is pushed forefront. A good algorithm can greatly increase happiness index people's lives. However, existing research field generally ignores relationship between attributes (such ignoring...
Point-of-Interest (POI) recommendation recommends different personalized services to interested users, which are widely used in people's daily life. However, with the massive increase users and POIs, POI system faces following challenging problems: (1) The results of service not enough, little attention is paid details semantic relevance. (2) Often face problem cold start (3) Recommended accuracy can be further improved. In order cope these difficulties, this paper proposes a method...
Personalized recipe recommendation is attracting more and attention, which can help people make choices from the exploding growth of online food information. Unlike other tasks, target a non-atomic item, so attribute information especially important for representation recipes. However, traditional collaborative filtering or content-based methods tend to focus on user–recipe interaction ignore higher-order semantic structural Recently, graph neural networks (GNNs)-based provided new ideas...
There are issues with the mining of association rules traditional Chinese medicine prescriptions, such as unclear rule strength, incomplete decision analysis, and low practicability. This paper introduces concept interest degree negative rules, well combining positive to analyze comprehensively. The strength is determined by correlation hidden in prescription. In this paper, an improved measure introduced, which effectively improves uneven value range commonly used measures, lacks zero...