Datasets:
shreysatapara
commited on
Commit
•
3e01be5
1
Parent(s):
fb5e2bd
Update README.md
Browse files
README.md
CHANGED
@@ -36,11 +36,58 @@ config_names:
|
|
36 |
- Hindi
|
37 |
- Gujarati
|
38 |
size_categories:
|
|
|
39 |
- 10K<n<100K
|
40 |
---
|
41 |
-
# Dataset Card for
|
42 |
|
43 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
- **Homepage:**
|
46 |
- **Repository:**
|
@@ -48,17 +95,13 @@ size_categories:
|
|
48 |
- **Leaderboard:**
|
49 |
- **Point of Contact:**
|
50 |
|
51 |
-
### Dataset Summary
|
52 |
|
53 |
-
This dataset card aims to be a base template for new datasets. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/datasetcard_template.md?plain=1).
|
54 |
|
55 |
### Supported Tasks and Leaderboards
|
56 |
|
57 |
[More Information Needed]
|
58 |
|
59 |
-
### Languages
|
60 |
|
61 |
-
[More Information Needed]
|
62 |
|
63 |
## Dataset Structure
|
64 |
|
@@ -66,17 +109,9 @@ This dataset card aims to be a base template for new datasets. It has been gener
|
|
66 |
|
67 |
[More Information Needed]
|
68 |
|
69 |
-
### Data Fields
|
70 |
-
|
71 |
-
[More Information Needed]
|
72 |
-
|
73 |
-
### Data Splits
|
74 |
|
75 |
-
[More Information Needed]
|
76 |
|
77 |
-
## Dataset Creation
|
78 |
|
79 |
-
### Curation Rationale
|
80 |
|
81 |
[More Information Needed]
|
82 |
|
@@ -128,10 +163,3 @@ This dataset card aims to be a base template for new datasets. It has been gener
|
|
128 |
|
129 |
[More Information Needed]
|
130 |
|
131 |
-
### Citation Information
|
132 |
-
|
133 |
-
[More Information Needed]
|
134 |
-
|
135 |
-
### Contributions
|
136 |
-
|
137 |
-
[More Information Needed]
|
|
|
36 |
- Hindi
|
37 |
- Gujarati
|
38 |
size_categories:
|
39 |
+
- 1K<n<10K
|
40 |
- 10K<n<100K
|
41 |
---
|
42 |
+
# Dataset Card for "ILSUM-1.0"
|
43 |
|
44 |
+
### Dataset Summary
|
45 |
+
|
46 |
+
Automatic text summarization for Indian languages has received surprisingly little attention from the NLP research community. While large scale datasets exist for a number of languages like English, Chinese, French, German, Spanish, etc. no such datasets exist for any Indian languages. Most existing datasets are either not public, or are too small to be useful. Through this shared task we aim to bridge the existing gap by creating reusable corpora for Indian Language Summarization. In the first edition we cover two major indian languages Hindi and Gujarati, which have over 350 million and over 50 million speakers respectively. Apart from this we also include Indian English, a widely regonized dialect which can be substantially different from English spoken elsewhere.
|
47 |
+
|
48 |
+
The dataset for this task is built using articles and headline pairs from several leading newspapers of the country. We provide ~10,000 news articles for each language. The task is to generate a meaningful fixed length summary, either extractive or abstractive, for each article. While several previous works in other languages use news artciles - headlines pair, the current dataset poses a unique challenge of code-mixing and script mixing. It is very common for news articles to borrow phrases from english, even if the article itself is written in an Indian Language.
|
49 |
+
|
50 |
+
Examples like these are a common occurence both in the headlines as well as in the articles.
|
51 |
+
~~~
|
52 |
+
- "IND vs SA, 5મી T20 તસવીરોમાં: વરસાદે વિલન બની મજા બગાડી" (India vs SA, 5th T20 in pictures: rain spoils the match)
|
53 |
+
- "LIC के IPO में पैसा लगाने वालों का टूटा दिल, आई एक और नुकसानदेह खबर" (Investors of LIC IPO left broken hearted, yet another bad news).
|
54 |
+
~~~
|
55 |
+
### Languages
|
56 |
+
- Hindi
|
57 |
+
- Gujarati
|
58 |
+
- English
|
59 |
+
|
60 |
+
### Data Fields
|
61 |
+
~~~
|
62 |
+
- id: Unique id of each datapoint
|
63 |
+
- Article: Entire News article
|
64 |
+
- Headline: Headline of News Article
|
65 |
+
- Summary: Summary of News Article
|
66 |
+
~~~
|
67 |
+
|
68 |
+
### Data Splits
|
69 |
+
Data for all three languages is divided into three splits train, validation and test.
|
70 |
+
|
71 |
+
|
72 |
+
### Citation Information
|
73 |
+
If you are using the dataset or the models please cite the following paper
|
74 |
+
~~~
|
75 |
+
@article{satapara2022findings,
|
76 |
+
title={Findings of the first shared task on indian language summarization (ilsum): Approaches, challenges and the path ahead},
|
77 |
+
author={Satapara, Shrey and Modha, Bhavan and Modha, Sandip and Mehta, Parth},
|
78 |
+
journal={Working Notes of FIRE},
|
79 |
+
pages={9--13},
|
80 |
+
year={2022}
|
81 |
+
}
|
82 |
+
~~~
|
83 |
+
|
84 |
+
|
85 |
+
### Contributions
|
86 |
+
- Bhavan Modha, University Of Texas at Dallas, USA
|
87 |
+
- Shrey Satapara, Indian Institute Of Technology, Hyderabad, India
|
88 |
+
- Sandip Modha, LDRP-ITR, Gandhinagar, India
|
89 |
+
- Parth Mehta, Parmonic, USA
|
90 |
+
<!--## Dataset Description
|
91 |
|
92 |
- **Homepage:**
|
93 |
- **Repository:**
|
|
|
95 |
- **Leaderboard:**
|
96 |
- **Point of Contact:**
|
97 |
|
|
|
98 |
|
|
|
99 |
|
100 |
### Supported Tasks and Leaderboards
|
101 |
|
102 |
[More Information Needed]
|
103 |
|
|
|
104 |
|
|
|
105 |
|
106 |
## Dataset Structure
|
107 |
|
|
|
109 |
|
110 |
[More Information Needed]
|
111 |
|
|
|
|
|
|
|
|
|
|
|
112 |
|
|
|
113 |
|
|
|
114 |
|
|
|
115 |
|
116 |
[More Information Needed]
|
117 |
|
|
|
163 |
|
164 |
[More Information Needed]
|
165 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|