Datasets:
Tasks:
Text Classification
Sub-tasks:
multi-class-classification
Multilinguality:
multilingual
Size Categories:
10K<n<100K
Language Creators:
found
Annotations Creators:
crowdsourced
Source Datasets:
original
ArXiv:
License:
Update README.md
Browse files
README.md
CHANGED
|
@@ -59,8 +59,9 @@ task_ids:
|
|
| 59 |
|
| 60 |
### Dataset Summary
|
| 61 |
|
| 62 |
-
|
| 63 |
|
|
|
|
| 64 |
Loading Russian Simple task:
|
| 65 |
```python
|
| 66 |
from datasets import load_dataset
|
|
@@ -69,6 +70,7 @@ dataset = load_dataset("IlyaGusev/headline_cause", "ru_simple")
|
|
| 69 |
|
| 70 |
### Supported Tasks and Leaderboards
|
| 71 |
|
|
|
|
| 72 |
|
| 73 |
### Languages
|
| 74 |
|
|
@@ -77,12 +79,40 @@ This dataset consists of two parts, Russian and English.
|
|
| 77 |
## Dataset Structure
|
| 78 |
|
| 79 |
### Data Instances
|
|
|
|
| 80 |
|
| 81 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
|
| 83 |
### Data Splits
|
| 84 |
|
| 85 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
|
| 87 |
## Dataset Creation
|
| 88 |
|
|
|
|
| 59 |
|
| 60 |
### Dataset Summary
|
| 61 |
|
| 62 |
+
A dataset for detecting implicit causal relations between pairs of news headlines. The dataset includes over 5000 headline pairs from English news and over 9000 headline pairs from Russian news labeled through crowdsourcing. The pairs vary from totally unrelated or belonging to the same general topic to the ones including causation and refutation relations.
|
| 63 |
|
| 64 |
+
### Usage
|
| 65 |
Loading Russian Simple task:
|
| 66 |
```python
|
| 67 |
from datasets import load_dataset
|
|
|
|
| 70 |
|
| 71 |
### Supported Tasks and Leaderboards
|
| 72 |
|
| 73 |
+
[More Information Needed]
|
| 74 |
|
| 75 |
### Languages
|
| 76 |
|
|
|
|
| 79 |
## Dataset Structure
|
| 80 |
|
| 81 |
### Data Instances
|
| 82 |
+
There is an URL, a title, and a timestamp for each of the two headlines in every data instance. A label is presented in three fields. 'Result' field is a textual label, 'label' field is a numeric label, and the 'agreement' field shows the majority vote agreement between annotators. Additional information includes instance ID and the presence of the link between two articles.
|
| 83 |
|
| 84 |
+
```
|
| 85 |
+
{
|
| 86 |
+
'left_url': 'https://www.kommersant.ru/doc/4347456',
|
| 87 |
+
'right_url': 'https://tass.ru/kosmos/8488527',
|
| 88 |
+
'left_title': 'NASA: информация об отказе сотрудничать с Россией по освоению Луны некорректна',
|
| 89 |
+
'right_title': 'NASA назвало некорректными сообщения о нежелании США включать РФ в соглашение по Луне',
|
| 90 |
+
'left_timestamp': datetime.datetime(2020, 5, 15, 19, 46, 20),
|
| 91 |
+
'right_timestamp': datetime.datetime(2020, 5, 15, 19, 21, 36),
|
| 92 |
+
'label': 0,
|
| 93 |
+
'result': 'not_cause',
|
| 94 |
+
'agreement': 1.0,
|
| 95 |
+
'id': 'ru_tg_101',
|
| 96 |
+
'has_link': True
|
| 97 |
+
}
|
| 98 |
+
```
|
| 99 |
|
| 100 |
### Data Splits
|
| 101 |
|
| 102 |
+
| Dataset | Split | Number of Instances |
|
| 103 |
+
|:---------|:---------|:---------|
|
| 104 |
+
| ru_simple | train | 7,641 |
|
| 105 |
+
| ru_simple | validation | 955 |
|
| 106 |
+
| ru_simple | test | 957 |
|
| 107 |
+
| en_simple | train | 4,332 |
|
| 108 |
+
| en_simple | validation | 542 |
|
| 109 |
+
| en_simple | test | 542 |
|
| 110 |
+
| ru_full | train | 5,713 |
|
| 111 |
+
| ru_full | validation | 715 |
|
| 112 |
+
| ru_full | test | 715 |
|
| 113 |
+
| en_full | train | 2,009 |
|
| 114 |
+
| en_full | validation | 251 |
|
| 115 |
+
| en_full | test | 252 |
|
| 116 |
|
| 117 |
## Dataset Creation
|
| 118 |
|