Datasets:
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,122 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
task_categories:
|
3 |
+
- token-classification
|
4 |
+
language:
|
5 |
+
- vi
|
6 |
+
---
|
7 |
+
## Dataset Card for BKEE
|
8 |
+
|
9 |
+
### 1. Dataset Summary
|
10 |
+
|
11 |
+
**BKEE** is the first Vietnamese **Event Extraction** dataset, containing **1,066** fully annotated documents covering:
|
12 |
+
|
13 |
+
* **33+ event types**
|
14 |
+
* **28 argument roles**
|
15 |
+
* Entity mentions, event mentions, and relation mentions
|
16 |
+
|
17 |
+
This resource addresses the lack of Vietnamese‑specific EE datasets and supports tasks from event detection to argument extraction.
|
18 |
+
|
19 |
+
|
20 |
+
### 2. Supported Tasks and Metrics
|
21 |
+
|
22 |
+
* **Tasks**
|
23 |
+
|
24 |
+
* Entity Mention Detection
|
25 |
+
* Event Detection
|
26 |
+
* Event Argument Extraction
|
27 |
+
* **Metrics**
|
28 |
+
|
29 |
+
* Precision / Recall / F1 (per subtask)
|
30 |
+
* Overall document‑level accuracy
|
31 |
+
|
32 |
+
|
33 |
+
### 3. Languages
|
34 |
+
|
35 |
+
* Vietnamese
|
36 |
+
|
37 |
+
### 4. Dataset Structure
|
38 |
+
|
39 |
+
All splits have been merged into one CSV. Each row corresponds to one sentence mention:
|
40 |
+
|
41 |
+
| Column | Type | Description |
|
42 |
+
| ------------------- | ------ | ----------------------------------------------------- |
|
43 |
+
| `doc_id` | string | Unique document identifier (e.g. “train-00001”). |
|
44 |
+
| `sent_id` | string | Unique sentence identifier within a document. |
|
45 |
+
| `tokens` | list | Tokenized words. |
|
46 |
+
| `sentence` | string | Original sentence text. |
|
47 |
+
| `event_types` | list | List of event type labels occurring in this sentence. |
|
48 |
+
| `argument_roles` | list | List of argument-role labels aligned to `tokens`. |
|
49 |
+
| `entity_mentions` | list | List of entity mention spans (start, end, label). |
|
50 |
+
| `event_mentions` | list | List of event trigger spans (start, end, label). |
|
51 |
+
| `relation_mentions` | list | List of relations between arguments and events. |
|
52 |
+
| `type` | string | Split: `train` / `dev` / `test`. |
|
53 |
+
| `dataset` | string | Always `BKEE` (for provenance). |
|
54 |
+
|
55 |
+
|
56 |
+
### 5. Data Fields
|
57 |
+
|
58 |
+
* **doc\_id**, **sent\_id** (`str`): Document and sentence IDs.
|
59 |
+
* **tokens** (`List[str]`): Tokenized text.
|
60 |
+
* **sentence** (`str`): Raw sentence.
|
61 |
+
* **event\_types**, **argument\_roles**, **entity\_mentions**, **event\_mentions**, **relation\_mentions**: Annotations per sentence.
|
62 |
+
* **type** (`str`): Data split.
|
63 |
+
* **dataset** (`str`): Always `BKEE`.
|
64 |
+
|
65 |
+
|
66 |
+
### 6. Usage
|
67 |
+
|
68 |
+
```python
|
69 |
+
from datasets import load_dataset
|
70 |
+
|
71 |
+
ds = load_dataset("visolex/BKEE")
|
72 |
+
|
73 |
+
# Filter by split
|
74 |
+
train = ds.filter(lambda ex: ex["type"] == "train")
|
75 |
+
val = ds.filter(lambda ex: ex["type"] == "dev")
|
76 |
+
test = ds.filter(lambda ex: ex["type"] == "test")
|
77 |
+
|
78 |
+
# Inspect one example
|
79 |
+
print(train[0])
|
80 |
+
```
|
81 |
+
|
82 |
+
|
83 |
+
### 7. Source & Links
|
84 |
+
|
85 |
+
* **GitHub** (original data & code):
|
86 |
+
[https://github.com/nhungnt/BKEE](https://github.com/nhungnt/BKEE) ([github.com][2])
|
87 |
+
* **Paper**:
|
88 |
+
Thi‑Nhung Nguyen et al. (2024), “BKEE: Pioneering Event Extraction in the Vietnamese Language,” *LREC‑COLING 2024*. ([aclanthology.org][1])
|
89 |
+
* **Hugging Face (this unified version)**:
|
90 |
+
[https://huggingface.co/datasets/visolex/BKEE](https://huggingface.co/datasets/visolex/BKEE)
|
91 |
+
|
92 |
+
|
93 |
+
### 8. License
|
94 |
+
|
95 |
+
Released under **CC BY‑NC 4.0**. See the GitHub repo for full terms.
|
96 |
+
|
97 |
+
|
98 |
+
### 9. Citation
|
99 |
+
|
100 |
+
```bibtex
|
101 |
+
@inproceedings{nguyen-etal-2024-bkee,
|
102 |
+
title = {BKEE: Pioneering Event Extraction in the Vietnamese Language},
|
103 |
+
author = {Nguyen, Thi-Nhung and Tran, Bang Tien and Luu, Trong-Nghia and
|
104 |
+
Nguyen, Thien Huu and Nguyen, Kiem-Hieu},
|
105 |
+
booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
|
106 |
+
month = may,
|
107 |
+
year = {2024},
|
108 |
+
address = {Torino, Italia},
|
109 |
+
publisher = {ELRA and ICCL},
|
110 |
+
url = {https://aclanthology.org/2024.lrec-main.217},
|
111 |
+
pages = {2421--2427}
|
112 |
+
}
|
113 |
+
```
|
114 |
+
|
115 |
+
```bibtex
|
116 |
+
@misc{nhungnt_bkee,
|
117 |
+
title = {BKEE: Pioneering Event Extraction in the Vietnamese Language},
|
118 |
+
author = {{nhungnt}},
|
119 |
+
howpublished = {\url{https://github.com/nhungnt/BKEE}},
|
120 |
+
year = {2024}
|
121 |
+
}
|
122 |
+
```
|