davanstrien HF staff commited on
Commit
f26037b
1 Parent(s): 88075f6

Create web_archive_classification.py

Browse files
Files changed (1) hide show
  1. web_archive_classification.py +208 -0
web_archive_classification.py ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Copyright 2022 The HuggingFace Datasets Authors and the current dataset script contributor.
2
+ #
3
+ # Licensed under the Apache License, Version 2.0 (the "License");
4
+ # you may not use this file except in compliance with the License.
5
+ # You may obtain a copy of the License at
6
+ #
7
+ # http://www.apache.org/licenses/LICENSE-2.0
8
+ #
9
+ # Unless required by applicable law or agreed to in writing, software
10
+ # distributed under the License is distributed on an "AS IS" BASIS,
11
+ # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12
+ # See the License for the specific language governing permissions and
13
+ # limitations under the License.
14
+ """British Library Web Classification Dataset."""
15
+
16
+ import datasets
17
+ import csv
18
+
19
+ _CITATION = """\
20
+ TODO
21
+ """
22
+
23
+ _DESCRIPTION = """\
24
+ The dataset comprises a manually curated selective archive produced by UKWA which includes the classification of sites into a two-tiered subject hierarchy.
25
+ """
26
+ _HOMEPAGE = "https://doi.org/10.5259/ukwa.ds.1/classification/1"
27
+
28
+ _LICENSE = "https://creativecommons.org/publicdomain/zero/1.0/"
29
+
30
+ _URL = "https://bl.iro.bl.uk/downloads/78e2421a-70ea-426d-8a67-57e4a8b23019?locale=en"
31
+
32
+
33
+ class WebArchiveClassificationDataset(datasets.GeneratorBasedBuilder):
34
+ """Web Archive Classification Dataset"""
35
+
36
+ VERSION = datasets.Version("1.1.0")
37
+
38
+ def _info(self):
39
+ features = datasets.Features(
40
+ {
41
+ "primary_category": datasets.ClassLabel(
42
+ names=[
43
+ "Arts & Humanities",
44
+ "Business, Economy & Industry",
45
+ "Company Web Sites",
46
+ "Computer Science, Information Technology and Web Technology",
47
+ "Crime, Criminology, Police and Prisons",
48
+ "Digital Society",
49
+ "Education & Research",
50
+ "Environment",
51
+ "Government, Law & Politics",
52
+ "History",
53
+ "Law and Legal System",
54
+ "Libraries, Archives and Museums",
55
+ "Life Sciences",
56
+ "Literature",
57
+ "Medicine & Health",
58
+ "Politics, Political Theory and Political Systems",
59
+ "Popular Science",
60
+ "Publishing, Printing and Bookselling",
61
+ "Religion",
62
+ "Science & Technology",
63
+ "Social Problems and Welfare",
64
+ "Society & Culture",
65
+ "Sports and Recreation",
66
+ "Travel & Tourism",
67
+ ]
68
+ ),
69
+ "secondary_category": datasets.ClassLabel(
70
+ names=[
71
+ "Architecture",
72
+ "Art and Design",
73
+ "Comedy and Humour",
74
+ "Dance",
75
+ "Family History / Genealogy",
76
+ "Film / Cinema",
77
+ "Geography",
78
+ "History",
79
+ "Languages",
80
+ "Literature",
81
+ "Live Art",
82
+ "Local History",
83
+ "Music",
84
+ "News and Contemporary Events",
85
+ "Oral History in the UK",
86
+ "Philosophy and Ethics",
87
+ "Publishing, Printing and Bookselling",
88
+ "Religion",
89
+ "TV and Radio",
90
+ "Theatre",
91
+ "Agriculture, Fishing, and Forestry",
92
+ "Banking, Insurance, Accountancy and Financial Economics",
93
+ "Business Studies and Management Theory",
94
+ "Company Web Sites",
95
+ "Credit Crunch",
96
+ "Economic Development, Enterprise and Aid",
97
+ "Economics and Economic Theory",
98
+ "Employment, Unemployment and Labour Economics",
99
+ "Energy",
100
+ "Industries",
101
+ "Marketing and Market Research",
102
+ "Trade, Commerce, and Globalisation",
103
+ "Transport and Infrastructure",
104
+ "Cambridge Network",
105
+ "Video Games",
106
+ "Governing the Police",
107
+ "Blogs",
108
+ "Dictionaries, Encyclopaedias, and Reference Works",
109
+ "Further Education",
110
+ "Higher Education",
111
+ "Libraries, Archives and Museums",
112
+ "Library Key Issues",
113
+ "Lifelong Learning",
114
+ "Preschool Education",
115
+ "School Education",
116
+ "Special Needs Education",
117
+ "Vocational Education",
118
+ "Indian Ocean Tsunami December 2004",
119
+ "Central Government",
120
+ "Civil Rights, Pressure Groups, and Trade Unions",
121
+ "Crime, Criminology, Police and Prisons",
122
+ "Devolved Government",
123
+ "European Parliament Elections 2009",
124
+ "Inter-Governmental Agencies",
125
+ "International Relations, Diplomacy, and Peace",
126
+ "Law and Legal System",
127
+ "Local Government",
128
+ "London Mayoral Election 2008",
129
+ "Political Parties",
130
+ "Politics, Political Theory and Political Systems",
131
+ "Public Inquiries",
132
+ "Scottish Parliamentary Election - 2007",
133
+ "Spending Cuts 2010: Impact on Social Welfare",
134
+ "UK General Election 2005",
135
+ "Slavery and Abolition in the Caribbean",
136
+ "Religion, politics and law since 2005",
137
+ "Evolving role of libraries in the UK",
138
+ "History of Libraries Collection",
139
+ "Darwin 200",
140
+ "19th Century English Literature",
141
+ "Alternative Medicine / Complementary Medicine",
142
+ "Conditions and Diseases",
143
+ "Health Organisations and Services",
144
+ "Medicines, Treatments and Therapies",
145
+ "Men's Issues",
146
+ "Mental Health",
147
+ "Pandemic Influenza",
148
+ "Personal Experiences of Illness",
149
+ "Public Health and Safety",
150
+ "Women's Issues",
151
+ "Political Action and Communication",
152
+ "E-publishing Trends",
153
+ "Free Church",
154
+ "Quakers",
155
+ "Computer Science, Information Technology and Web Technology",
156
+ "Engineering",
157
+ "Environment",
158
+ "Life Sciences",
159
+ "Mathematics",
160
+ "Physical Sciences",
161
+ "Popular Science",
162
+ "Zoology, Veterinary Science and Animal Health",
163
+ "Communities",
164
+ "Digital Society",
165
+ "Food and Drink",
166
+ "London Terrorist Attack 7th July 2005",
167
+ "Queen's Diamond Jubilee, 2012",
168
+ "Social Problems and Welfare",
169
+ "Sociology, Anthropology and Population Studies",
170
+ "Sports and Recreation",
171
+ "Travel & Tourism",
172
+ "British Countryside",
173
+ "Olympic & Paralympic Games 2012",
174
+ "Cornwall",
175
+ ]
176
+ ),
177
+ "title": datasets.Value("string"),
178
+ "url": datasets.Value("string"),
179
+ }
180
+ )
181
+ return datasets.DatasetInfo(
182
+ description=_DESCRIPTION,
183
+ features=features,
184
+ homepage=_HOMEPAGE,
185
+ license=_LICENSE,
186
+ citation=_CITATION,
187
+ )
188
+
189
+ def _split_generators(self, dl_manager):
190
+
191
+ csv_file = dl_manager.download_and_extract(_URL)
192
+ return [
193
+ datasets.SplitGenerator(
194
+ name=datasets.Split.TRAIN,
195
+ gen_kwargs={"csv_file": csv_file},
196
+ ),
197
+ ]
198
+
199
+ def _generate_examples(self, csv_file):
200
+ with open(csv_file) as f:
201
+ reader = csv.DictReader(f, dialect="excel-tab")
202
+ for id_, row in enumerate(reader):
203
+ yield id_, {
204
+ "primary_category": row["Primary Category"],
205
+ "secondary_category": row["Secondary Category"],
206
+ "title": row["Title"],
207
+ "url": row["URL"],
208
+ }