Research map / Graph learning / Knowledge and temporal graphs
TransConv: Relationship Embedding in Social Networks
Yi-Yu Lai, Jennifer Neville, Dan Goldwasser
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add conversation similarity and frequency factors to TransH score function (TransConv vs TransH) (ablation) | TransConv · Facebook link prediction | Hits@10 filtered (%): 76.3 -> 86.9 | Improved |
| add conversation factors to TransH (TransConv vs TransH) (ablation) | TransConv · Twitter link prediction | Hits@10 filtered (%): 92.4 -> 98.3 | Improved |
| add conversation factors to TransH (ablation) | TransConv · Facebook triplet classification (replace tail) | accuracy (%): 71.2 -> 94.9 | Improved |
| add conversation factors to TransH (ablation) | TransConv · Twitter triplet classification (replace tail) | accuracy (%): 65.4 -> 99.9 | Improved |
| add conversation factors to TransH (ablation) | TransConv · Facebook multilabel classification | Hamming accuracy (%): 13.8 -> 15.2 | Improved |
| set alpha to 0 or 1 (use only one conversation factor) instead of 0.5 (ablation) | TransConv · Facebook link prediction | link prediction performance | Hurt |
| conversation factors on sparse (bottom-3) relationships | TransConv · Facebook bottom-3 relationships | Hits@10: TransConv nearly over 60% vs others below 20% | Improved |
| conversation factors on top-3 frequent relationships | TransConv · Facebook top-3 relationships | Hits@10 | No clear effect |
| use user-message keyword descriptions (DKRL) instead of structure only | TransE · Facebook link prediction | Hits@10 filtered (%): 52.3 -> 6.7 | Hurt |
Later papers that compared with this method
1 later accepted paper reports Trans.Conv as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Graph Neural Networks Need Cluster-Normalize-Activate Modules (NeurIPS 2024 · better on 2 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Convolutional 2D Knowledge Graph Embeddings (AAAI 2018 · similarity 0.87)
- TransGate: Knowledge Graph Embedding with Shared Gate Structure (AAAI 2019 · similarity 0.87)
- TorusE: Knowledge Graph Embedding on a Lie Group (AAAI 2018 · similarity 0.86)
- Locally Adaptive Translation for Knowledge Graph Embedding (AAAI 2016 · similarity 0.85)
- Graph Structure Learning on User Mobility Data for Social Relationship Inference (AAAI 2023 · similarity 0.85)
- Non-Euclidean Mixture Model for Social Network Embedding (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
