Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,56 @@
|
|
1 |
---
|
2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
+
task_categories:
|
4 |
+
- text-generation
|
5 |
+
pretty_name: KoWoW
|
6 |
+
size_categories:
|
7 |
+
- 100K<n<1M
|
8 |
+
language:
|
9 |
+
- ko
|
10 |
+
tags:
|
11 |
+
- chat
|
12 |
+
- mutiturn conversation
|
13 |
+
- chatbot
|
14 |
---
|
15 |
+
# Dataset Card for KoWoW
|
16 |
+
|
17 |
+
<!-- Provide a quick summary of the dataset. -->
|
18 |
+
|
19 |
+
|
20 |
+
## Dataset Summary
|
21 |
+
|
22 |
+
WoW(Wiard of Wikipedia)를 한국어로 변역한 데이터입니다.
|
23 |
+
|
24 |
+
|
25 |
+
### Dataset Description
|
26 |
+
WoW(Wiard of Wikipedia)라는 지식 기반 대화 데이터를 한국어로 변역한 데이터입니다.
|
27 |
+
한 대화에 여러 개의 dialog가 묶음으로 구성되어 있으며, 전체 대화는 22,311건, 전체 dialog는 201,999개 입니다.
|
28 |
+
본 데이터셋은 Knowledge와 Utterance가 모두 한국어인 ko 버전만 가져온 데이터입니다.
|
29 |
+
|
30 |
+
|
31 |
+
- **Language(s) (NLP):** ko
|
32 |
+
- **License:** mit
|
33 |
+
|
34 |
+
### Dataset Sources
|
35 |
+
|
36 |
+
<!-- Provide the basic links for the dataset. -->
|
37 |
+
|
38 |
+
- **Repository:** https://github.com/AIRC-KETI/kowow/tree/master
|
39 |
+
|
40 |
+
|
41 |
+
## Uses
|
42 |
+
|
43 |
+
<!-- Address questions around how the dataset is intended to be used. -->
|
44 |
+
|
45 |
+
|
46 |
+
### Source Data
|
47 |
+
|
48 |
+
<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->
|
49 |
+
WoW(Wiard of Wikipedia)
|
50 |
+
|
51 |
+
|
52 |
+
#### Who are the source data producers?
|
53 |
+
|
54 |
+
<!-- This section describes the people or systems who originally created the data. It should also include self-reported demographic or identity information for the source data creators if this information is available. -->
|
55 |
+
|
56 |
+
- AIRC-KETI
|