Datasets:
Upload folder using huggingface_hub
Browse files- README.md +94 -0
- train.jsonl.gz +3 -0
README.md
ADDED
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license: cc-by-sa-3.0
|
5 |
+
task_categories:
|
6 |
+
- text-classification
|
7 |
+
- summarization
|
8 |
+
- topic-modeling
|
9 |
+
- text-generation
|
10 |
+
task_ids:
|
11 |
+
- multi-class-classification
|
12 |
+
- abstractive-summarization
|
13 |
+
- topic-classification
|
14 |
+
- language-modeling
|
15 |
+
pretty_name: Wikipedia Paragraph Samples
|
16 |
+
tags:
|
17 |
+
- wikipedia
|
18 |
+
- paragraphs
|
19 |
+
- english
|
20 |
+
dataset_info:
|
21 |
+
features:
|
22 |
+
- name: title
|
23 |
+
dtype: string
|
24 |
+
- name: text
|
25 |
+
dtype: string
|
26 |
+
---
|
27 |
+
|
28 |
+
# Wikipedia Paragraph Samples
|
29 |
+
|
30 |
+
## Dataset Description
|
31 |
+
|
32 |
+
This dataset contains paragraphs extracted from randomly selected English Wikipedia articles. It provides a diverse sample of Wikipedia content across various topics.
|
33 |
+
|
34 |
+
## Dataset Details
|
35 |
+
|
36 |
+
- **Name**: Wikipedia Paragraph Samples
|
37 |
+
- **Version**: 1.0
|
38 |
+
- **Date Created**: 2024-08-20
|
39 |
+
- **Language**: English
|
40 |
+
- **Format**: JSONLines
|
41 |
+
|
42 |
+
## Contents
|
43 |
+
|
44 |
+
Each line in the dataset represents a single paragraph and contains two fields:
|
45 |
+
1. Title of the Wikipedia page
|
46 |
+
2. Text of the paragraph
|
47 |
+
|
48 |
+
## Size and Scope
|
49 |
+
|
50 |
+
- **Number of Samples**: 5329
|
51 |
+
- **Paragraph Length**: 1000 to 3000 characters
|
52 |
+
|
53 |
+
## Data Collection Process
|
54 |
+
|
55 |
+
- **Source**: English Wikipedia
|
56 |
+
- **Selection Method**: Random sampling of articles
|
57 |
+
- **Extraction Date**: 2024-08-20
|
58 |
+
|
59 |
+
## Data Format
|
60 |
+
|
61 |
+
Example entry:
|
62 |
+
|
63 |
+
```json
|
64 |
+
{
|
65 |
+
"title": "Cubs\\u2013White Sox rivalry",
|
66 |
+
"text": "The rivalry turned physical on May 20, 2006, when a brawl broke out during a White Sox-Cubs game at U.S. Cellular Field. In the bottom of the second inning, Brian Anderson of the White Sox hit a sacrifice fly, attempting to score catcher A. J. Pierzynski. Pierzynski collided with Cubs' catcher Michael Barrett. Barrett dropped the ball in the collision and Pierzynski was safe. After slapping home plate in celebration, Pierzynski began to walk away, but Barrett blocked his path and punched him in the jaw. Both benches cleared and a brawl broke out. Umpires debated for 15 minutes before ejecting Pierzynski, Barrett, White Sox outfielder Brian Anderson and Cubs first baseman John Mabry from the game. When play finally resumed, outfielder Scott Podsednik promptly got on base, loading the bases up, and second baseman Tadahito Iguchi cleared them with a grand-slam. The White Sox won the game, 7\\u20130. Michael Barrett was suspended for 10 games, while Brian Anderson was suspended for five and A. J. Pierzynski was fined."
|
67 |
+
}
|
68 |
+
```
|
69 |
+
|
70 |
+
## Intended Uses
|
71 |
+
|
72 |
+
This dataset can be useful for various natural language processing tasks, including but not limited to:
|
73 |
+
- Text classification
|
74 |
+
- Topic modeling
|
75 |
+
- Text summarization
|
76 |
+
- Language model fine-tuning
|
77 |
+
- Information retrieval research
|
78 |
+
|
79 |
+
## Limitations and Biases
|
80 |
+
|
81 |
+
- Content is limited to topics covered in Wikipedia
|
82 |
+
- May contain biases present in Wikipedia's coverage and editing practices
|
83 |
+
- Randomly selected articles may not provide balanced representation across all topics
|
84 |
+
|
85 |
+
## Ethical Considerations
|
86 |
+
|
87 |
+
- Users should be aware of Wikipedia's content policies and potential biases
|
88 |
+
- Proper attribution should be given when using this data
|
89 |
+
- Care should be taken when using for applications that may impact real-world decisions
|
90 |
+
|
91 |
+
## Additional Notes
|
92 |
+
|
93 |
+
- This dataset provides a snapshot of Wikipedia content as of the extraction date
|
94 |
+
- Users should refer to Wikipedia's terms of use for guidelines on using its content
|
train.jsonl.gz
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6260045162bc82b35e021b32786ece17b0281db3deb94ea1f78283f3c152b458
|
3 |
+
size 2722670
|