Datasets:
Commit
•
5027966
0
Parent(s):
Update files from the datasets library (from 1.2.0)
Browse filesRelease notes: https://github.com/huggingface/datasets/releases/tag/1.2.0
- .gitattributes +27 -0
- README.md +211 -0
- dataset_infos.json +1 -0
- dummy/dutch_social/1.1.0/dummy_data.zip +3 -0
- dutch_social.py +213 -0
.gitattributes
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bin.* filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
20 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
26 |
+
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,211 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
annotations_creators:
|
3 |
+
- machine-generated
|
4 |
+
language_creators:
|
5 |
+
- crowdsourced
|
6 |
+
languages:
|
7 |
+
- en
|
8 |
+
- nl
|
9 |
+
licenses:
|
10 |
+
- cc-by-nc-4-0
|
11 |
+
multilinguality:
|
12 |
+
- multilingual
|
13 |
+
size_categories:
|
14 |
+
- 100K< n<1M
|
15 |
+
source_datasets:
|
16 |
+
- original
|
17 |
+
task_categories:
|
18 |
+
- text-classification
|
19 |
+
task_ids:
|
20 |
+
- sentiment-classification
|
21 |
+
- multi-label-classification
|
22 |
+
---
|
23 |
+
|
24 |
+
# Dataset Card for Dutch Social Media Collection
|
25 |
+
|
26 |
+
## Table of Contents
|
27 |
+
- [Dataset Description](#dataset-description)
|
28 |
+
- [Dataset Summary](#dataset-summary)
|
29 |
+
- [Supported Tasks](#supported-tasks-and-leaderboards)
|
30 |
+
- [Languages](#languages)
|
31 |
+
- [Dataset Structure](#dataset-structure)
|
32 |
+
- [Data Instances](#data-instances)
|
33 |
+
- [Data Fields](#data-instances)
|
34 |
+
- [Data Splits](#data-instances)
|
35 |
+
- [Dataset Creation](#dataset-creation)
|
36 |
+
- [Curation Rationale](#curation-rationale)
|
37 |
+
- [Source Data](#source-data)
|
38 |
+
- [Annotations](#annotations)
|
39 |
+
- [Personal and Sensitive Information](#personal-and-sensitive-information)
|
40 |
+
- [Considerations for Using the Data](#considerations-for-using-the-data)
|
41 |
+
- [Social Impact of Dataset](#social-impact-of-dataset)
|
42 |
+
- [Discussion of Biases](#discussion-of-biases)
|
43 |
+
- [Other Known Limitations](#other-known-limitations)
|
44 |
+
- [Additional Information](#additional-information)
|
45 |
+
- [Dataset Curators](#dataset-curators)
|
46 |
+
- [Licensing Information](#licensing-information)
|
47 |
+
- [Citation Information](#citation-information)
|
48 |
+
|
49 |
+
## Dataset Description
|
50 |
+
|
51 |
+
- **Homepage:[Dutch Social Media Collection](http://datasets.coronawhy.org/dataset.xhtml?persistentId=doi:10.5072/FK2/MTPTL7)**
|
52 |
+
- **Repository: **
|
53 |
+
- **Paper:*(in-progress)* https://doi.org/10.5072/FK2/MTPTL7**
|
54 |
+
- **Leaderboard:**
|
55 |
+
- **Point of Contact: [Aakash Gupta](mailto:aakashg80@gmail.com)**
|
56 |
+
|
57 |
+
### Dataset Summary
|
58 |
+
|
59 |
+
The dataset contains 10 files with around 271,342 tweets. The tweets are filtered via the official Twitter API to contain tweets in Dutch language or by users who have specified their location information within Netherlands geographical boundaries. Using natural language processing we have classified the tweets for their HISCO codes. If the user has provided their location within Dutch boundaries, we have also classified them to their respective provinces The objective of this dataset is to make research data available publicly in a FAIR (Findable, Accessible, Interoperable, Reusable) way. Twitter's Terms of Service Licensed under Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) (2020-10-27)
|
60 |
+
|
61 |
+
### Supported Tasks and Leaderboards
|
62 |
+
|
63 |
+
`sentiment analysis`, `multi-label classification`, `entity-extraction`
|
64 |
+
|
65 |
+
### Languages
|
66 |
+
|
67 |
+
The text is primarily in Dutch with some tweets in English and other languages. The BCP 47 code is `nl` and `en`
|
68 |
+
|
69 |
+
## Dataset Structure
|
70 |
+
|
71 |
+
### Data Instances
|
72 |
+
|
73 |
+
An example of the data field will be:
|
74 |
+
|
75 |
+
```
|
76 |
+
{
|
77 |
+
"full_text": "@pflegearzt @Friedelkorn @LAguja44 Pardon, wollte eigentlich das zitieren: \nhttps://t.co/ejO7bIMyj8\nMeine mentions sind inzw komplett undurchschaubar weil da Leute ihren supporterclub zwecks Likes zusammengerufen haben.",
|
78 |
+
"text_translation": "@pflegearzt @Friedelkorn @ LAguja44 Pardon wollte zitieren eigentlich das:\nhttps://t.co/ejO7bIMyj8\nMeine mentions inzw sind komplett undurchschaubar weil da Leute ihren supporter club Zwecks Likes zusammengerufen haben.",
|
79 |
+
"created_at": 1583756789000,
|
80 |
+
"screen_name": "TheoRettich",
|
81 |
+
"description": "I ❤️science, therefore a Commie. ☭ FALGSC: Part of a conspiracy which wants to achieve world domination. Tankie-Cornucopian. Ecology is a myth",
|
82 |
+
"desc_translation": "I ❤️science, Therefore a Commie. ☭ FALGSC: Part of a conspiracy How many followers wants to Achieve World Domination. Tankie-Cornucopian. Ecology is a myth",
|
83 |
+
"weekofyear": 11,
|
84 |
+
"weekday": 0,
|
85 |
+
"day": 9,
|
86 |
+
"month": 3,
|
87 |
+
"year": 2020,
|
88 |
+
"location": "Netherlands",
|
89 |
+
"point_info": "Nederland",
|
90 |
+
"point": "(52.5001698, 5.7480821, 0.0)",
|
91 |
+
"latitude": 52.5001698,
|
92 |
+
"longitude": 5.7480821,
|
93 |
+
"altitude": 0,
|
94 |
+
"province": "Flevoland",
|
95 |
+
"hisco_standard": null,
|
96 |
+
"hisco_code": null,
|
97 |
+
"industry": false,
|
98 |
+
"sentiment_pattern": 0,
|
99 |
+
"subjective_pattern": 0
|
100 |
+
}
|
101 |
+
```
|
102 |
+
|
103 |
+
### Data Fields
|
104 |
+
|
105 |
+
|
106 |
+
| Column Name | Description |
|
107 |
+
| --- | --- |
|
108 |
+
| full_text | Original text in the tweet |
|
109 |
+
| text_translation | English translation of the full text |
|
110 |
+
| created_at | Date of tweet creation |
|
111 |
+
| screen_name | username of the tweet author |
|
112 |
+
| description | description as provided in the users bio |
|
113 |
+
| desc_translation | English translation of user's bio/ description |
|
114 |
+
| location | Location information as provided in the user's bio |
|
115 |
+
| weekofyear | week of the year |
|
116 |
+
| weekday | Day of the week information; Monday=0....Sunday = 6|
|
117 |
+
| month | Month of tweet creation |
|
118 |
+
| year | year of tweet creation |
|
119 |
+
| day | day of tweet creation |
|
120 |
+
| point_info | point information from location columnd |
|
121 |
+
| point | tuple giving lat, lon & altitude information |
|
122 |
+
| latitude | geo-referencing information derived from location data |
|
123 |
+
| longitude | geo-referencing information derived from location data |
|
124 |
+
| altitude | geo-referencing information derived from location data|
|
125 |
+
| province | Province given location data of user |
|
126 |
+
| hisco_standard | HISCO standard key word; if available in tweet |
|
127 |
+
| hisco_code| HISCO standard code as derived from `hisco_standard`|
|
128 |
+
| industry | Whether the tweet talks about industry `(True/False)` |
|
129 |
+
| sentiment_score | Sentiment score -1.0 to 1.0 |
|
130 |
+
| subjectivity_score | Subjectivity scores 0 to 1 |
|
131 |
+
|
132 |
+
Missing values are replaced with empty strings or -1 (-100 for missing sentiment_score).
|
133 |
+
|
134 |
+
|
135 |
+
### Data Splits
|
136 |
+
|
137 |
+
Data has been split into Train: 60%, Validation: 20% and Test: 20%
|
138 |
+
|
139 |
+
## Dataset Creation
|
140 |
+
|
141 |
+
### Curation Rationale
|
142 |
+
|
143 |
+
[More Information Needed]
|
144 |
+
|
145 |
+
### Source Data
|
146 |
+
|
147 |
+
#### Initial Data Collection and Normalization
|
148 |
+
|
149 |
+
The tweets were hydrated using Twitter's API and then filtered for those which were in Dutch language and/or for users who had mentioned that they were from within Netherlands geographical borders.
|
150 |
+
|
151 |
+
#### Who are the source language producers?
|
152 |
+
|
153 |
+
The language producers are twitter users who have identified their location within the geographical boundaries of Netherland. Or those who have tweeted in the dutch language!
|
154 |
+
|
155 |
+
### Annotations
|
156 |
+
|
157 |
+
Using Natural language processing, we have classified the tweets on industry and for HSN HISCO codes.
|
158 |
+
Depending on the user's location, their provincial information is also added. Please check the file/column for detailed information.
|
159 |
+
|
160 |
+
The tweets are also classified on the sentiment & subjectivity scores.
|
161 |
+
Sentiment scores are between -1 to +1
|
162 |
+
Subjectivity scores are between 0 to 1
|
163 |
+
|
164 |
+
#### Annotation process
|
165 |
+
|
166 |
+
[More Information Needed]
|
167 |
+
|
168 |
+
#### Who are the annotators?
|
169 |
+
|
170 |
+
[More Information Needed]
|
171 |
+
|
172 |
+
### Personal and Sensitive Information
|
173 |
+
|
174 |
+
As of writing this data card no anonymization has been carried out on the tweets or user data. As such, if the twitter user has shared any personal & sensitive information, then it may be available in this dataset.
|
175 |
+
|
176 |
+
## Considerations for Using the Data
|
177 |
+
|
178 |
+
### Social Impact of Dataset
|
179 |
+
|
180 |
+
[More Information Needed]
|
181 |
+
|
182 |
+
### Discussion of Biases
|
183 |
+
|
184 |
+
[More Information Needed]
|
185 |
+
|
186 |
+
### Other Known Limitations
|
187 |
+
|
188 |
+
[More Information Needed]
|
189 |
+
|
190 |
+
## Additional Information
|
191 |
+
|
192 |
+
### Dataset Curators
|
193 |
+
|
194 |
+
[Aakash Gupta](mailto:aakashg80@gmail.com)
|
195 |
+
*Th!nkEvolve Consulting* and Researcher at CoronaWhy
|
196 |
+
|
197 |
+
### Licensing Information
|
198 |
+
|
199 |
+
CC BY-NC 4.0
|
200 |
+
|
201 |
+
### Citation Information
|
202 |
+
|
203 |
+
@data{FK2/MTPTL7_2020,
|
204 |
+
author = {Gupta, Aakash},
|
205 |
+
publisher = {COVID-19 Data Hub},
|
206 |
+
title = {{Dutch social media collection}},
|
207 |
+
year = {2020},
|
208 |
+
version = {DRAFT VERSION},
|
209 |
+
doi = {10.5072/FK2/MTPTL7},
|
210 |
+
url = {https://doi.org/10.5072/FK2/MTPTL7}
|
211 |
+
}
|
dataset_infos.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"dutch_social": {"description": "The dataset contains around 271,342 tweets. The tweets are filtered via the official Twitter API to\ncontain tweets in Dutch language or by users who have specified their location information within Netherlands\ngeographical boundaries. Using natural language processing we have classified the tweets for their HISCO codes.\nIf the user has provided their location within Dutch boundaries, we have also classified them to their respective\nprovinces The objective of this dataset is to make research data available publicly in a FAIR (Findable, Accessible,\nInteroperable, Reusable) way. Twitter's Terms of Service Licensed under Attribution-NonCommercial 4.0 International\n(CC BY-NC 4.0) (2020-10-27)\n", "citation": "@data{FK2/MTPTL7_2020,\nauthor = {Gupta, Aakash},\npublisher = {COVID-19 Data Hub},\ntitle = {{Dutch social media collection}},\nyear = {2020},\nversion = {DRAFT VERSION},\ndoi = {10.5072/FK2/MTPTL7},\nurl = {https://doi.org/10.5072/FK2/MTPTL7}\n}\n", "homepage": "http://datasets.coronawhy.org/dataset.xhtml?persistentId=doi:10.5072/FK2/MTPTL7", "license": "CC BY-NC 4.0", "features": {"full_text": {"dtype": "string", "id": null, "_type": "Value"}, "text_translation": {"dtype": "string", "id": null, "_type": "Value"}, "screen_name": {"dtype": "string", "id": null, "_type": "Value"}, "description": {"dtype": "string", "id": null, "_type": "Value"}, "desc_translation": {"dtype": "string", "id": null, "_type": "Value"}, "location": {"dtype": "string", "id": null, "_type": "Value"}, "weekofyear": {"dtype": "int64", "id": null, "_type": "Value"}, "weekday": {"dtype": "int64", "id": null, "_type": "Value"}, "month": {"dtype": "int64", "id": null, "_type": "Value"}, "year": {"dtype": "int64", "id": null, "_type": "Value"}, "day": {"dtype": "int64", "id": null, "_type": "Value"}, "point_info": {"dtype": "string", "id": null, "_type": "Value"}, "point": {"dtype": "string", "id": null, "_type": "Value"}, "latitude": {"dtype": "float64", "id": null, "_type": "Value"}, "longitude": {"dtype": "float64", "id": null, "_type": "Value"}, "altitude": {"dtype": "float64", "id": null, "_type": "Value"}, "province": {"dtype": "string", "id": null, "_type": "Value"}, "hisco_standard": {"dtype": "string", "id": null, "_type": "Value"}, "hisco_code": {"dtype": "string", "id": null, "_type": "Value"}, "industry": {"dtype": "bool_", "id": null, "_type": "Value"}, "sentiment_pattern": {"dtype": "float64", "id": null, "_type": "Value"}, "subjective_pattern": {"dtype": "float64", "id": null, "_type": "Value"}, "label": {"num_classes": 3, "names": ["neg", "neu", "pos"], "names_file": null, "id": null, "_type": "ClassLabel"}}, "post_processed": null, "supervised_keys": null, "builder_name": "dutch_social", "config_name": "dutch_social", "version": {"version_str": "1.1.0", "description": null, "major": 1, "minor": 1, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 105569586, "num_examples": 162805, "dataset_name": "dutch_social"}, "test": {"name": "test", "num_bytes": 35185351, "num_examples": 54268, "dataset_name": "dutch_social"}, "validation": {"name": "validation", "num_bytes": 34334756, "num_examples": 54269, "dataset_name": "dutch_social"}}, "download_checksums": {"https://storage.googleapis.com/corona-tweet/dutch-tweets.zip": {"num_bytes": 68740666, "checksum": "29a080692e806d5f05011b3157ae2c91a7964c32cc90c7f30532ab6dc980053a"}}, "download_size": 68740666, "post_processing_size": null, "dataset_size": 175089693, "size_in_bytes": 243830359}}
|
dummy/dutch_social/1.1.0/dummy_data.zip
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:770b3209c87309cc010a56c7243515c83a45296c27f15d8b1ad5edb5cce739a8
|
3 |
+
size 4570
|
dutch_social.py
ADDED
@@ -0,0 +1,213 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# coding=utf-8
|
2 |
+
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
|
3 |
+
#
|
4 |
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
5 |
+
# you may not use this file except in compliance with the License.
|
6 |
+
# You may obtain a copy of the License at
|
7 |
+
#
|
8 |
+
# http://www.apache.org/licenses/LICENSE-2.0
|
9 |
+
#
|
10 |
+
# Unless required by applicable law or agreed to in writing, software
|
11 |
+
# distributed under the License is distributed on an "AS IS" BASIS,
|
12 |
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
13 |
+
# See the License for the specific language governing permissions and
|
14 |
+
# limitations under the License.
|
15 |
+
"""DUTCH SOCIAL: Annotated Covid19 tweets in Dutch language (sentiment, industry codes & province)."""
|
16 |
+
|
17 |
+
from __future__ import absolute_import, division, print_function
|
18 |
+
|
19 |
+
import json
|
20 |
+
import os
|
21 |
+
|
22 |
+
import datasets
|
23 |
+
|
24 |
+
|
25 |
+
# TODO: Add BibTeX citation
|
26 |
+
# Find for instance the citation on arxiv or on the dataset repo/website
|
27 |
+
_CITATION = """\
|
28 |
+
@data{FK2/MTPTL7_2020,
|
29 |
+
author = {Gupta, Aakash},
|
30 |
+
publisher = {COVID-19 Data Hub},
|
31 |
+
title = {{Dutch social media collection}},
|
32 |
+
year = {2020},
|
33 |
+
version = {DRAFT VERSION},
|
34 |
+
doi = {10.5072/FK2/MTPTL7},
|
35 |
+
url = {https://doi.org/10.5072/FK2/MTPTL7}
|
36 |
+
}
|
37 |
+
"""
|
38 |
+
|
39 |
+
# TODO: Add description of the dataset here
|
40 |
+
# You can copy an official description
|
41 |
+
_DESCRIPTION = """\
|
42 |
+
The dataset contains around 271,342 tweets. The tweets are filtered via the official Twitter API to
|
43 |
+
contain tweets in Dutch language or by users who have specified their location information within Netherlands
|
44 |
+
geographical boundaries. Using natural language processing we have classified the tweets for their HISCO codes.
|
45 |
+
If the user has provided their location within Dutch boundaries, we have also classified them to their respective
|
46 |
+
provinces The objective of this dataset is to make research data available publicly in a FAIR (Findable, Accessible,
|
47 |
+
Interoperable, Reusable) way. Twitter's Terms of Service Licensed under Attribution-NonCommercial 4.0 International
|
48 |
+
(CC BY-NC 4.0) (2020-10-27)
|
49 |
+
"""
|
50 |
+
|
51 |
+
# TODO: Add a link to an official homepage for the dataset here
|
52 |
+
_HOMEPAGE = "http://datasets.coronawhy.org/dataset.xhtml?persistentId=doi:10.5072/FK2/MTPTL7"
|
53 |
+
|
54 |
+
# TODO: Add the licence for the dataset here if you can find it
|
55 |
+
_LICENSE = "CC BY-NC 4.0"
|
56 |
+
|
57 |
+
# TODO: Add link to the official dataset URLs here
|
58 |
+
# The HuggingFace dataset library don't host the datasets but only point to the original files
|
59 |
+
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
|
60 |
+
_URLs = {"dutch_social": "https://storage.googleapis.com/corona-tweet/dutch-tweets.zip"}
|
61 |
+
|
62 |
+
_LANG = ["nl", "en"]
|
63 |
+
|
64 |
+
|
65 |
+
# TODO: Name of the dataset usually match the script name with CamelCase instead of snake_case
|
66 |
+
class DutchSocial(datasets.GeneratorBasedBuilder):
|
67 |
+
"""
|
68 |
+
Annotated Covid19 tweets in Dutch language. The tweets were filtered for users who had indicated
|
69 |
+
their location within Netherlands or if the tweets were in Dutch language. The purpose of curating
|
70 |
+
these tweets is to measure the economic impact of the Covid19 pandemic
|
71 |
+
"""
|
72 |
+
|
73 |
+
VERSION = datasets.Version("1.1.0")
|
74 |
+
|
75 |
+
# This is an example of a dataset with multiple configurations.
|
76 |
+
# If you don't want/need to define several sub-sets in your dataset,
|
77 |
+
# just remove the BUILDER_CONFIG_CLASS and the BUILDER_CONFIGS attributes.
|
78 |
+
|
79 |
+
# If you need to make complex sub-parts in the datasets with configurable options
|
80 |
+
# You can create your own builder configuration class to store attribute, inheriting from datasets.BuilderConfig
|
81 |
+
# BUILDER_CONFIG_CLASS = MyBuilderConfig
|
82 |
+
|
83 |
+
# You will be able to load one or the other configurations in the following list with
|
84 |
+
# data = datasets.load_dataset('my_dataset', 'first_domain')
|
85 |
+
# data = datasets.load_dataset('my_dataset', 'second_domain')
|
86 |
+
BUILDER_CONFIGS = [
|
87 |
+
datasets.BuilderConfig(
|
88 |
+
name="dutch_social",
|
89 |
+
version=VERSION,
|
90 |
+
description="This part of my dataset provides config for the entire dataset",
|
91 |
+
)
|
92 |
+
# datasets.BuilderConfig(name="second_domain", version=VERSION, description="This part of my dataset covers a second domain"),
|
93 |
+
]
|
94 |
+
|
95 |
+
def _info(self):
|
96 |
+
# TODO: This method specifies the datasets.DatasetInfo object which contains informations and typings for the dataset
|
97 |
+
features = datasets.Features(
|
98 |
+
{
|
99 |
+
"full_text": datasets.Value("string"),
|
100 |
+
"text_translation": datasets.Value("string"),
|
101 |
+
"screen_name": datasets.Value("string"),
|
102 |
+
"description": datasets.Value("string"),
|
103 |
+
"desc_translation": datasets.Value("string"),
|
104 |
+
"location": datasets.Value("string"),
|
105 |
+
"weekofyear": datasets.Value("int64"),
|
106 |
+
"weekday": datasets.Value("int64"),
|
107 |
+
"month": datasets.Value("int64"),
|
108 |
+
"year": datasets.Value("int64"),
|
109 |
+
"day": datasets.Value("int64"),
|
110 |
+
"point_info": datasets.Value("string"),
|
111 |
+
"point": datasets.Value("string"),
|
112 |
+
"latitude": datasets.Value("float64"),
|
113 |
+
"longitude": datasets.Value("float64"),
|
114 |
+
"altitude": datasets.Value("float64"),
|
115 |
+
"province": datasets.Value("string"),
|
116 |
+
"hisco_standard": datasets.Value("string"),
|
117 |
+
"hisco_code": datasets.Value("string"),
|
118 |
+
"industry": datasets.Value("bool_"),
|
119 |
+
"sentiment_pattern": datasets.Value("float64"),
|
120 |
+
"subjective_pattern": datasets.Value("float64"),
|
121 |
+
"label": datasets.ClassLabel(num_classes=3, names=["neg", "neu", "pos"], names_file=None, id=None),
|
122 |
+
}
|
123 |
+
)
|
124 |
+
return datasets.DatasetInfo(
|
125 |
+
# This is the description that will appear on the datasets page.
|
126 |
+
description=_DESCRIPTION,
|
127 |
+
# This defines the different columns of the dataset and their types
|
128 |
+
features=features, # Here we define them above because they are different between the two configurations
|
129 |
+
# If there's a common (input, target) tuple from the features,
|
130 |
+
# specify them here. They'll be used if as_supervised=True in
|
131 |
+
# builder.as_dataset.
|
132 |
+
supervised_keys=None,
|
133 |
+
# Homepage of the dataset for documentation
|
134 |
+
homepage=_HOMEPAGE,
|
135 |
+
# License for the dataset if available
|
136 |
+
license=_LICENSE,
|
137 |
+
# Citation for the dataset
|
138 |
+
citation=_CITATION,
|
139 |
+
)
|
140 |
+
|
141 |
+
def _split_generators(self, dl_manager):
|
142 |
+
"""Returns SplitGenerators."""
|
143 |
+
# TODO: This method is tasked with downloading/extracting the data and defining the splits depending on the configuration
|
144 |
+
# If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name
|
145 |
+
|
146 |
+
# dl_manager is a datasets.download.DownloadManager that can be used to download and extract URLs
|
147 |
+
# It can accept any type or nested list/dict and will give back the same structure with the url replaced with path to local files.
|
148 |
+
# By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
|
149 |
+
my_urls = _URLs[self.config.name]
|
150 |
+
data_dir = dl_manager.download_and_extract(my_urls)
|
151 |
+
|
152 |
+
return [
|
153 |
+
datasets.SplitGenerator(
|
154 |
+
name=datasets.Split.TRAIN,
|
155 |
+
# These kwargs will be passed to _generate_examples
|
156 |
+
gen_kwargs={
|
157 |
+
"filepath": os.path.join(data_dir, "train.jsonl"),
|
158 |
+
"split": "train",
|
159 |
+
},
|
160 |
+
),
|
161 |
+
datasets.SplitGenerator(
|
162 |
+
name=datasets.Split.TEST,
|
163 |
+
# These kwargs will be passed to _generate_examples
|
164 |
+
gen_kwargs={"filepath": os.path.join(data_dir, "test.jsonl"), "split": "test"},
|
165 |
+
),
|
166 |
+
datasets.SplitGenerator(
|
167 |
+
name=datasets.Split.VALIDATION,
|
168 |
+
# These kwargs will be passed to _generate_examples
|
169 |
+
gen_kwargs={
|
170 |
+
"filepath": os.path.join(data_dir, "dev.jsonl"),
|
171 |
+
"split": "dev",
|
172 |
+
},
|
173 |
+
),
|
174 |
+
]
|
175 |
+
|
176 |
+
def _generate_examples(self, filepath, split, key=None):
|
177 |
+
""" Yields examples. """
|
178 |
+
# TODO: This method will receive as arguments the `gen_kwargs` defined in the previous `_split_generators` method.
|
179 |
+
# It is in charge of opening the given file and yielding (key, example) tuples from the dataset
|
180 |
+
# The key is not important, it's more here for legacy reason (legacy from tfds)
|
181 |
+
|
182 |
+
with open(filepath, encoding="utf-8") as f:
|
183 |
+
for id_, data in enumerate(f):
|
184 |
+
data = json.loads(data)
|
185 |
+
yield id_, {
|
186 |
+
"full_text": "" if not isinstance(data["full_text"], str) else data["full_text"],
|
187 |
+
"text_translation": ""
|
188 |
+
if not isinstance(data["text_translation"], str)
|
189 |
+
else data["text_translation"],
|
190 |
+
"screen_name": "" if not isinstance(data["screen_name"], str) else data["screen_name"],
|
191 |
+
"description": "" if not isinstance(data["description"], str) else data["description"],
|
192 |
+
"desc_translation": ""
|
193 |
+
if not isinstance(data["desc_translation"], str)
|
194 |
+
else data["desc_translation"],
|
195 |
+
"location": "" if not isinstance(data["location"], str) else data["location"],
|
196 |
+
"weekofyear": -1 if data["weekofyear"] is None else data["weekofyear"],
|
197 |
+
"weekday": -1 if data["weekday"] is None else data["weekday"],
|
198 |
+
"month": -1 if data["month"] is None else data["month"],
|
199 |
+
"year": -1 if data["year"] is None else data["year"],
|
200 |
+
"day": -1 if data["day"] is None else data["day"],
|
201 |
+
"point_info": "" if isinstance(data["point_info"], str) else data["point_info"],
|
202 |
+
"point": "" if not isinstance(data["point"], str) else data["point"],
|
203 |
+
"latitude": -1 if data["latitude"] is None else data["latitude"],
|
204 |
+
"longitude": -1 if data["longitude"] is None else data["longitude"],
|
205 |
+
"altitude": -1 if data["altitude"] is None else data["altitude"],
|
206 |
+
"province": "" if not isinstance(data["province"], str) else data["province"],
|
207 |
+
"hisco_standard": "" if not isinstance(data["hisco_standard"], str) else data["hisco_standard"],
|
208 |
+
"hisco_code": "" if not isinstance(data["hisco_code"], str) else data["hisco_code"],
|
209 |
+
"industry": False if not isinstance(data["industry"], bool) else data["industry"],
|
210 |
+
"sentiment_pattern": -100 if data["sentiment_pattern"] is None else data["sentiment_pattern"],
|
211 |
+
"subjective_pattern": -1 if data["subjective_pattern"] is None else data["subjective_pattern"],
|
212 |
+
"label": data["label"],
|
213 |
+
}
|