Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -33,4 +33,58 @@ tags:
|
|
| 33 |
- medical
|
| 34 |
size_categories:
|
| 35 |
- 10K<n<100K
|
| 36 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
- medical
|
| 34 |
size_categories:
|
| 35 |
- 10K<n<100K
|
| 36 |
+
---
|
| 37 |
+
|
| 38 |
+
# KorMedMCQA : Multi-Choice Question Answering Benchmark for Korean Healthcare Professional Licensing Examinations
|
| 39 |
+
|
| 40 |
+
We introduce KorMedMCQA, the first Korean multiple-choice question answering (MCQA) benchmark derived from Korean healthcare professional licensing examinations, covering from the year 2012 to year 2023.
|
| 41 |
+
This dataset consists of a selection of questions from the license examinations for doctors, nurses, and pharmacists, featuring a diverse array of subjects.
|
| 42 |
+
We conduct baseline experiments on various large language models, including proprietary/open-source, multilingual/Korean-additional pretrained, and clinical context pretrained models, highlighting the potential for further enhancements.
|
| 43 |
+
We make our data publicly available on HuggingFace and provide a evaluation script via LM-Harness, inviting further exploration and advancement in Korean healthcare environments.
|
| 44 |
+
|
| 45 |
+
Paper : https://arxiv.org/abs/2403.01469
|
| 46 |
+
|
| 47 |
+
## Dataset Details
|
| 48 |
+
|
| 49 |
+
### Languages
|
| 50 |
+
|
| 51 |
+
Korean
|
| 52 |
+
|
| 53 |
+
### Subtask
|
| 54 |
+
|
| 55 |
+
```
|
| 56 |
+
from datasets import load_dataset
|
| 57 |
+
doctor = load_dataset(path = "sean0042/KorMedMCQA",name = "doctor")
|
| 58 |
+
nurse = load_dataset(path = "sean0042/KorMedMCQA",name = "nurse")
|
| 59 |
+
pharmacist = load_dataset(path = "sean0042/KorMedMCQA",name = "pharm")
|
| 60 |
+
```
|
| 61 |
+
|
| 62 |
+
### Statistics
|
| 63 |
+
|
| 64 |
+
| Category | # Questions (Train/Dev/Test) |
|
| 65 |
+
|------------------------------|------------------------------|
|
| 66 |
+
| Doctor | 2,339 (1,890/164/285) |
|
| 67 |
+
| Nurse | 1,460 (582/291/587) |
|
| 68 |
+
| Pharmacist | 1,546 (632/300/614) |
|
| 69 |
+
|
| 70 |
+
### Data Fields
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
- `subject`: doctor, nurse, or pharm
|
| 74 |
+
- `year`: year of the examination
|
| 75 |
+
- `period`: period of the examination
|
| 76 |
+
- `q_number`: question number of the examination
|
| 77 |
+
- `question`: question
|
| 78 |
+
- `A`: First answer choice
|
| 79 |
+
- `B`: Second answer choice
|
| 80 |
+
- `C`: Third answer choice
|
| 81 |
+
- `D`: Fourth answer choice
|
| 82 |
+
- `E`: Fifth answer choice
|
| 83 |
+
- `answer` : Answer (1 to 5). 1 denotes answer A, and 5 denotes answer E
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
## Contact
|
| 87 |
+
|
| 88 |
+
```
|
| 89 |
+
sean0042@kaist.ac.kr
|
| 90 |
+
```
|