KG-Hi-BKF Benchmark for Biomedical Knowledge Fusion

The KG-Hi-BKF benchmark datasets for the biomedical knowledge integration task, which is proposed in our SIGIR'2023 paper: "HiPrompt: Few-Shot Biomedical Knowledge Fusion via Hierarchy-Oriented Prompting". <br> There are two datasets: SDKG-DzHi repoDB-DzHi <br> Under each dataset, the files are organized as follows: ent_attr_KG.json ent_attr_Hi.json rel_trip_KG.csv hypo_hyper_pair_Hi.csv ent_links.csv zero_shot/ |-- test_ent_links.csv one_shot/ |-- support_ent_links.csv |-- test_ent_links.csv <br> where `ent_attr*.json` denotes entity attribute files, `rel_trip_KG.csv` denotes knowledge graph relational triples, `hypo_hyper_pair_Hi.csv` denotes hierarchy hyponym-hypernym term pairs, and `ent_links.csv` denotes all aligned entity-term pairs. The two subfolders `zero_shot/` and `one_shot/` provide the support&amp;test split used in the paper. <br> If you use the datasets, please cite our paper ```BibTex @inproceedings{lu23HiPrompt,<br> title = {HiPrompt: Few-Shot Biomedical Knowledge Fusion via Hierarchy-Oriented Prompting},<br> author = {Lu, Jiaying and Shen, Jiaming and Xiong, Bo and Ma, Wengjing and Staab Steffen and Yang, Carl},<br> year = {2023},<br> month = {Apr.},<br> Series = {SIGIR 2023},<br> Booktitle = {46th International ACM SIGIR Conference on Research and Development in Information Retrieval - Short Paper},<br> } ```本数据集为面向生物医学知识融合（Biomedical Knowledge Integration）任务的KG-Hi-BKF基准数据集，相关研究成果发表于我们2023年SIGIR会议的论文《HiPrompt：面向层级提示的少样本（Few-Shot）生物医学知识融合》（HiPrompt: Few-Shot Biomedical Knowledge Fusion via Hierarchy-Oriented Prompting）。本次提供的数据集包含两个子数据集：SDKG-DzHi与repoDB-DzHi。每个子数据集下的文件组织形式如下： 包含实体属性（Entity Attribute）文件`ent_attr_KG.json`、`ent_attr_Hi.json`，知识图谱（Knowledge Graph）关系三元组（Relational Triples）文件`rel_trip_KG.csv`，层级上下位词对（Hyponym-Hypernym Term Pair）文件`hypo_hyper_pair_Hi.csv`，以及实体对齐配对文件`ent_links.csv`；另有两个子目录： - `zero_shot/`（零样本（Zero-Shot）划分）：仅包含测试集（Test Set）文件`test_ent_links.csv` - `one_shot/`（单样本（One-Shot）划分）：包含支持集（Support Set）文件`support_ent_links.csv`与测试集（Test Set）文件`test_ent_links.csv` 其中，`ent_attr*.json`指代实体属性文件，`rel_trip_KG.csv`指代知识图谱关系三元组文件，`hypo_hyper_pair_Hi.csv`指代层级上下位术语对文件，`ent_links.csv`指代所有对齐的实体-术语配对文件。`zero_shot/`与`one_shot/`两个子文件夹分别提供了本论文中使用的数据集划分方案。若您在研究中使用本数据集，请引用如下论文： @inproceedings{lu23HiPrompt, title = {HiPrompt: Few-Shot Biomedical Knowledge Fusion via Hierarchy-Oriented Prompting}, author = {Lu, Jiaying and Shen, Jiaming and Xiong, Bo and Ma, Wengjing and Staab Steffen and Yang, Carl}, year = {2023}, month = {Apr.}, Series = {SIGIR 2023}, Booktitle = {46th International ACM SIGIR Conference on Research and Development in Information Retrieval - Short Paper}, }