Enhancing Link Prediction with Fuzzy Graph Attention Networks and Dynamic Negative Sampling
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912214109126656 |
|---|---|
| author | Xing, Jinming Xing, Ruilin Xue, Chang Luo, Dongwen |
| author_facet | Xing, Jinming Xing, Ruilin Xue, Chang Luo, Dongwen |
| contents | Link prediction is crucial for understanding complex networks but traditional Graph Neural Networks (GNNs) often rely on random negative sampling, leading to suboptimal performance. This paper introduces Fuzzy Graph Attention Networks (FGAT), a novel approach integrating fuzzy rough sets for dynamic negative sampling and enhanced node feature aggregation. Fuzzy Negative Sampling (FNS) systematically selects high-quality negative edges based on fuzzy similarities, improving training efficiency. FGAT layer incorporates fuzzy rough set principles, enabling robust and discriminative node representations. Experiments on two research collaboration networks demonstrate FGAT's superior link prediction accuracy, outperforming state-of-the-art baselines by leveraging the power of fuzzy rough sets for effective negative sampling and node feature learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_07482 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing Link Prediction with Fuzzy Graph Attention Networks and Dynamic Negative Sampling Xing, Jinming Xing, Ruilin Xue, Chang Luo, Dongwen Machine Learning Artificial Intelligence Information Retrieval Link prediction is crucial for understanding complex networks but traditional Graph Neural Networks (GNNs) often rely on random negative sampling, leading to suboptimal performance. This paper introduces Fuzzy Graph Attention Networks (FGAT), a novel approach integrating fuzzy rough sets for dynamic negative sampling and enhanced node feature aggregation. Fuzzy Negative Sampling (FNS) systematically selects high-quality negative edges based on fuzzy similarities, improving training efficiency. FGAT layer incorporates fuzzy rough set principles, enabling robust and discriminative node representations. Experiments on two research collaboration networks demonstrate FGAT's superior link prediction accuracy, outperforming state-of-the-art baselines by leveraging the power of fuzzy rough sets for effective negative sampling and node feature learning. |
| title | Enhancing Link Prediction with Fuzzy Graph Attention Networks and Dynamic Negative Sampling |
| topic | Machine Learning Artificial Intelligence Information Retrieval |
| url | https://arxiv.org/abs/2411.07482 |