Strong Transitivity Relations and Graph Neural Networks
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_ | 1866913184258981888 |
|---|---|
| author | Mohamadi, Yassin Chehreghani, Mostafa Haghir |
| author_facet | Mohamadi, Yassin Chehreghani, Mostafa Haghir |
| contents | Local neighborhoods play a crucial role in embedding generation in graph-based learning. It is commonly believed that nodes ought to have embeddings that resemble those of their neighbors. In this research, we try to carefully expand the concept of similarity from nearby neighborhoods to the entire graph. We provide an extension of similarity that is based on transitivity relations, which enables Graph Neural Networks (GNNs) to capture both global similarities and local similarities over the whole graph. We introduce Transitivity Graph Neural Network (TransGNN), which more than local node similarities, takes into account global similarities by distinguishing strong transitivity relations from weak ones and exploiting them. We evaluate our model over several real-world datasets and showed that it considerably improves the performance of several well-known GNN models, for tasks such as node classification. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_01384 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Strong Transitivity Relations and Graph Neural Networks Mohamadi, Yassin Chehreghani, Mostafa Haghir Social and Information Networks Artificial Intelligence Machine Learning Local neighborhoods play a crucial role in embedding generation in graph-based learning. It is commonly believed that nodes ought to have embeddings that resemble those of their neighbors. In this research, we try to carefully expand the concept of similarity from nearby neighborhoods to the entire graph. We provide an extension of similarity that is based on transitivity relations, which enables Graph Neural Networks (GNNs) to capture both global similarities and local similarities over the whole graph. We introduce Transitivity Graph Neural Network (TransGNN), which more than local node similarities, takes into account global similarities by distinguishing strong transitivity relations from weak ones and exploiting them. We evaluate our model over several real-world datasets and showed that it considerably improves the performance of several well-known GNN models, for tasks such as node classification. |
| title | Strong Transitivity Relations and Graph Neural Networks |
| topic | Social and Information Networks Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2401.01384 |