KS-GNN: Keywords Search over Incomplete Graphs via Graphs Neural Network
Authors: YU HAO, Xin Cao, Yufan Sheng, Yixiang Fang, Wei Wang
NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The experiments on four real-world datasets show that our model consistently achieves better performance than state-of-the-art baseline methods in graphs having missing information. |
| Researcher Affiliation | Academia | Yu Hao University of New South Wales NSW, Australia yu.hao@unsw.edu.au Xin Cao University of New South Wales NSW, Australia xin.cao@unsw.edu.au Yufan Sheng University of New South Wales NSW, Australia yufan.sheng@unsw.edu.au Yixiang Fang Chinese University of Hong Kong, Shenzhen, China fangyixiang@cuhk.edu.cn Wei Wang The Hong Kong University of Science and Technology, Guangzhou, China weiwcs@ust.hk |
| Pseudocode | No | The paper describes the model's mechanisms and equations but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes] |
| Open Datasets | Yes | We evaluate the performance of our proposed approach, KS-GNN, on four real-world datasets, including citation networks (Cite Seer), co-purchase networks (Video & Toy) and co-author networks (DBLP). In Section 4.a of the checklist, it states: If your work uses existing assets, did you cite the creators? [Yes] See Appendix B. |
| Dataset Splits | Yes | In each incomplete graph, the validation set consists of 100 randomly generated queries with ground truth answers. We tune the hyper-parameters of compared methods with the grid search algorithm on the validation set, more details can be found in the supplementary materials. |
| Hardware Specification | Yes | Did you include the total amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)? [Yes] See Appendix C.1. |
| Software Dependencies | No | The paper mentions various models and frameworks (e.g., GNN, MLP, Graph SAGE, BLINK, SAT) but does not specify particular software library names with version numbers required for replication. |
| Experiment Setup | No | Section 5.2 'Experimental Setup' states: 'We tune the hyper-parameters of compared methods with the grid search algorithm on the validation set, more details can be found in the supplementary materials.' This indicates the specific details are not in the main text. |