Research map / Graph learning / Graph algorithms and clustering
KS-GNN: Keywords Search over Incomplete Graphs via Graphs Neural Network
YU HAO, Xin Cao, Yufan SHENG, Yixiang Fang, Wei Wang
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 L1 objective (ablation) | KS-GNN · CiteSeer, rw=0.3, nq=3 | Hits@100: 28.17 -> 30.84 | Improved |
| add L1 objective (ablation) | KS-GNN · Video | Hits@100: 17.82 -> 21.43 | Improved |
| add L1 objective (ablation) | KS-GNN · Toy | Hits@100: 24.67 -> 28.56 | Improved |
| add L1 objective (ablation) | KS-GNN · DBLP | Hits@100: 14.62 -> 16.21 | Improved |
| add L2 objective (ablation) | KS-GNN · Video | Hits@100: 18.86 -> 21.43 | Improved |
| add L3 objective (ablation) | KS-GNN · CiteSeer | Hits@100: 26.28 -> 30.84 | Improved |
| add L3 objective (ablation) | KS-GNN · Toy | Hits@100: 27.01 -> 28.56 | Hurt |
| average gain of L1, L2, L3 across four datasets (ablation) | KS-GNN · four datasets | Hits@100: 12.23%, 6.87%, 10.32% | Improved |
| replace PCA transformations with MLP encoder/decoder (KS-PCA to KS-GNN) (ablation) | KS-PCA · Video, rw=0.3, nq=3 | Hits@100: 10.19 -> 21.43 | Improved |
| add proposed message passing (Conv-PCA to KS-PCA) (ablation) | Conv-PCA · CiteSeer, rw=0.3, nq=3 | Hits@100: 8.49 -> 24.91 | Improved |
| add missing edges (re=0.3) | KS-GNN · Video, rw=0.3, nq=3 | Hits@100: 21.43 -> 8.08 | Hurt |
| add missing edges (re=0.3) | KS-GNN · CiteSeer, rw=0.3, nq=3 | Hits@100: 30.84 -> 30.57 | No clear effect |
Closest papers by meaning
- KerGNNs: Interpretable Graph Neural Networks with Graph Kernels (AAAI 2022 · similarity 0.86)
- INDIGO: GNN-Based Inductive Knowledge Graph Completion Using Pair-Wise Encoding (NeurIPS 2021 · similarity 0.86)
- Mixed Geometry Message and Trainable Convolutional Attention Network for Knowledge Graph Completion (AAAI 2024 · similarity 0.84)
- Graph Neural Prompting with Large Language Models (AAAI 2024 · similarity 0.84)
- RAGraph: A General Retrieval-Augmented Graph Learning Framework (NeurIPS 2024 · similarity 0.83)
- Diffusing to the Top: Boost Graph Neural Networks with Minimal Hyperparameter Tuning (ICLR 2025 · similarity 0.83)
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.
