Datasets:
Merge branch 'main' of https://huggingface.co/datasets/clips/mqa into main
Browse files
README.md
CHANGED
@@ -60,9 +60,9 @@ task_ids:
|
|
60 |
---
|
61 |
# MQA
|
62 |
MQA is a Multilingual corpus of Questions and Answers (MQA) parsed from the [Common Crawl](https://commoncrawl.org/). Questions are divided in two types: *Frequently Asked Questions (FAQ)* and *Community Question Answering (CQA)*.
|
63 |
-
```
|
64 |
from datasets import load_dataset
|
65 |
-
|
66 |
{
|
67 |
"name": "the title of the question (if any)",
|
68 |
"text": "the body of the question (if any)",
|
@@ -71,13 +71,13 @@ all = load_dataset("clips/mqa", language="en")
|
|
71 |
"is_accepted": "true|false"
|
72 |
}]
|
73 |
}
|
74 |
-
|
75 |
-
|
76 |
```
|
77 |
|
78 |
## Languages
|
79 |
We collected around **234M pairs** of questions and answers in **39 languages**. To download a language specific subset you need to specify the language key as configuration. See below for an example.
|
80 |
-
```
|
81 |
load_dataset("clips/mqa", language="en") # replace "en" by any language listed below
|
82 |
```
|
83 |
|
@@ -126,7 +126,7 @@ load_dataset("clips/mqa", language="en") # replace "en" by any language listed b
|
|
126 |
## FAQ vs. CQA
|
127 |
You can download the *Frequently Asked Questions* (FAQ) or the *Community Question Answering* (CQA) part of the dataset.
|
128 |
|
129 |
-
```
|
130 |
faq = load_dataset("clips/mqa", scope="faq")
|
131 |
cqa = load_dataset("clips/mqa", scope="cqa")
|
132 |
all = load_dataset("clips/mqa", scope="all")
|
@@ -136,27 +136,27 @@ Although FAQ and CQA questions share the same structure, CQA questions can have
|
|
136 |
## Nesting and Data Fields
|
137 |
You can specify three different nesting level: `question`, `page` and `domain`.
|
138 |
#### Question
|
139 |
-
```
|
140 |
load_dataset("clips/mqa", level="question") # default
|
141 |
```
|
142 |
The default level is the question object:
|
143 |
-
- **name**: the title of the question
|
144 |
-
- **text**: the body of the question (if any)
|
145 |
- **answers**: a list of answers
|
146 |
-
- **text**: the title of the answer (if any)
|
147 |
-
- **name**: the body of the answer
|
148 |
- **is_accepted**: true if the answer is selected.
|
149 |
|
150 |
#### Page
|
151 |
This level returns a list of questions present on the same page. This is mostly useful for FAQs since CQAs already have one question per page.
|
152 |
-
```
|
153 |
load_dataset("clips/mqa", level="page")
|
154 |
```
|
155 |
|
156 |
#### Domain
|
157 |
This level returns a list of pages present on the web domain. This is a good way to cope with FAQs duplication by sampling one page per domain at each epoch.
|
158 |
|
159 |
-
```
|
160 |
load_dataset("clips/mqa", level="domain")
|
161 |
```
|
162 |
|
@@ -187,10 +187,19 @@ We will comply to legitimate requests by removing the affected sources from the
|
|
187 |
|
188 |
## Citation information
|
189 |
```
|
190 |
-
@
|
191 |
-
|
192 |
-
|
193 |
-
|
194 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
195 |
}
|
|
|
196 |
```
|
|
|
60 |
---
|
61 |
# MQA
|
62 |
MQA is a Multilingual corpus of Questions and Answers (MQA) parsed from the [Common Crawl](https://commoncrawl.org/). Questions are divided in two types: *Frequently Asked Questions (FAQ)* and *Community Question Answering (CQA)*.
|
63 |
+
```python
|
64 |
from datasets import load_dataset
|
65 |
+
all_data = load_dataset("clips/mqa", language="en")
|
66 |
{
|
67 |
"name": "the title of the question (if any)",
|
68 |
"text": "the body of the question (if any)",
|
|
|
71 |
"is_accepted": "true|false"
|
72 |
}]
|
73 |
}
|
74 |
+
faq_data = load_dataset("clips/mqa", scope="faq", language="en")
|
75 |
+
cqa_data = load_dataset("clips/mqa", scope="cqa", language="en")
|
76 |
```
|
77 |
|
78 |
## Languages
|
79 |
We collected around **234M pairs** of questions and answers in **39 languages**. To download a language specific subset you need to specify the language key as configuration. See below for an example.
|
80 |
+
```python
|
81 |
load_dataset("clips/mqa", language="en") # replace "en" by any language listed below
|
82 |
```
|
83 |
|
|
|
126 |
## FAQ vs. CQA
|
127 |
You can download the *Frequently Asked Questions* (FAQ) or the *Community Question Answering* (CQA) part of the dataset.
|
128 |
|
129 |
+
```python
|
130 |
faq = load_dataset("clips/mqa", scope="faq")
|
131 |
cqa = load_dataset("clips/mqa", scope="cqa")
|
132 |
all = load_dataset("clips/mqa", scope="all")
|
|
|
136 |
## Nesting and Data Fields
|
137 |
You can specify three different nesting level: `question`, `page` and `domain`.
|
138 |
#### Question
|
139 |
+
```python
|
140 |
load_dataset("clips/mqa", level="question") # default
|
141 |
```
|
142 |
The default level is the question object:
|
143 |
+
- **name**: the title of the question(if any) in markdown format
|
144 |
+
- **text**: the body of the question (if any) in markdown format
|
145 |
- **answers**: a list of answers
|
146 |
+
- **text**: the title of the answer (if any) in markdown format
|
147 |
+
- **name**: the body of the answer in markdown format
|
148 |
- **is_accepted**: true if the answer is selected.
|
149 |
|
150 |
#### Page
|
151 |
This level returns a list of questions present on the same page. This is mostly useful for FAQs since CQAs already have one question per page.
|
152 |
+
```python
|
153 |
load_dataset("clips/mqa", level="page")
|
154 |
```
|
155 |
|
156 |
#### Domain
|
157 |
This level returns a list of pages present on the web domain. This is a good way to cope with FAQs duplication by sampling one page per domain at each epoch.
|
158 |
|
159 |
+
```python
|
160 |
load_dataset("clips/mqa", level="domain")
|
161 |
```
|
162 |
|
|
|
187 |
|
188 |
## Citation information
|
189 |
```
|
190 |
+
@inproceedings{de-bruyn-etal-2021-mfaq,
|
191 |
+
title = "{MFAQ}: a Multilingual {FAQ} Dataset",
|
192 |
+
author = "De Bruyn, Maxime and
|
193 |
+
Lotfi, Ehsan and
|
194 |
+
Buhmann, Jeska and
|
195 |
+
Daelemans, Walter",
|
196 |
+
booktitle = "Proceedings of the 3rd Workshop on Machine Reading for Question Answering",
|
197 |
+
month = nov,
|
198 |
+
year = "2021",
|
199 |
+
address = "Punta Cana, Dominican Republic",
|
200 |
+
publisher = "Association for Computational Linguistics",
|
201 |
+
url = "https://aclanthology.org/2021.mrqa-1.1",
|
202 |
+
pages = "1--13",
|
203 |
}
|
204 |
+
|
205 |
```
|