Hugging Face's logo

Datasets:
recogna-nlp
/
recognasumm

Tasks:
Summarization
Languages:
Portuguese
Size:
100K<n<1M
Tags:
pt
pt-br
summarization
abstractive summarization
news
License:
Dataset card Viewer Files Community
1
recognasumm
File size: 2,664 Bytes 
0322b81
 
1f9335f
 
 
 
 
 
 
 
 
 
 
 
 
0322b81
1f9335f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
d995568
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
 
---
license: mit
task_categories:
- summarization
language:
- pt
tags:
- pt
- pt-br
- summarization
- abstractive summarization
- news
pretty_name: RecognaSumm
size_categories:
- 100K<n<1M
---

# RecognaSumm Dataset

## Introduction
RecognaSumm is a novel and comprehensive database specifically designed for the task of automatic text summarization in Portuguese. RecognaSumm stands out due to its diverse origin, composed of news collected from a variety of information sources, including agencies and online news portals. The database was constructed using web scraping techniques and careful curation, re sulting in a rich and representative collection of documents covering various topics and journalis tic styles. The creation of RecognaSumm aims to fill a significant void in Portuguese language summarization research, providing a training and evaluation foundation that can be used for the development and enhancement of automated summarization models.

## News Categories
| Category                                                                                                | # of news| 
| :-:                                                                                                    | :-:                        | 
|Brazil | 14,131 |
|Economy | 12,613 |
|Entertainment | 5,337|
|Health | 24,921| 
|Policy | 29,909 |
|Science and Technology | 15,135 |
|Sports | 2,915 |
|Travel and Gastronomy | 2,893 |
| World | 27,418 |
| **Total** | **135,272** |

## PTT5-Summ Model

We also trained the [PTT5](https://github.com/unicamp-dl/PTT5) model on this dataset and made it available on HuggingFace. [Click here to access](https://huggingface.co/recogna-nlp/ptt5-base-summ).

# Citation

### RecognaSumm: A Novel Brazilian Summarization Dataset (PROPOR 2024)

```
@inproceedings{paiola-etal-2024-recognasumm,
    title = "{R}ecogna{S}umm: A Novel {B}razilian Summarization Dataset",
    author = "Paiola, Pedro Henrique  and
      Garcia, Gabriel Lino  and
      Jodas, Danilo Samuel  and
      Correia, Jo{\~a}o Vitor Mariano  and
      Sugi, Luis Afonso  and
      Papa, Jo{\~a}o Paulo",
    editor = "Gamallo, Pablo  and
      Claro, Daniela  and
      Teixeira, Ant{\'o}nio  and
      Real, Livy  and
      Garcia, Marcos  and
      Oliveira, Hugo Gon{\c{c}}alo  and
      Amaro, Raquel",
    booktitle = "Proceedings of the 16th International Conference on Computational Processing of Portuguese - Vol. 1",
    month = mar,
    year = "2024",
    address = "Santiago de Compostela, Galicia/Spain",
    publisher = "Association for Computational Lingustics",
    url = "https://aclanthology.org/2024.propor-1.63",
    pages = "575--579",
}
```