Datasets:

Languages:
English
Multilinguality:
monolingual
Size Categories:
1K<n<10K
Language Creators:
found
Annotations Creators:
found
Source Datasets:
original
Tags:
License:
File size: 13,506 Bytes
81f0d09
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
99a2f49
81f0d09
 
 
 
 
 
 
 
 
 
 
c2e6273
81f0d09
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
81f0d09
097ef55
81f0d09
 
 
 
 
 
097ef55
81f0d09
097ef55
81f0d09
 
 
 
 
 
097ef55
81f0d09
097ef55
81f0d09
 
 
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
097ef55
 
81f0d09
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c2e6273
81f0d09
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
96f5248
81f0d09
 
 
 
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
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
# coding=utf-8
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""This is an authorship attribution dataset based on the work of Stamatatos 2013. """


import os

import datasets


_CITATION = """\
@article{article,
    author = {Stamatatos, Efstathios},
    year = {2013},
    month = {01},
    pages = {421-439},
    title = {On the robustness of authorship attribution based on character n-gram features},
    volume = {21},
    journal = {Journal of Law and Policy}
}

@inproceedings{stamatatos2017authorship,
    title={Authorship attribution using text distortion},
    author={Stamatatos, Efstathios},
    booktitle={Proc. of the 15th Conf. of the European Chapter of the Association for Computational Linguistics},
    volume={1}
    pages={1138--1149},
    year={2017}
}
"""

_DESCRIPTION = """\
A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013.
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows.
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>",
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>",
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')

IMPORTANT: train+validation+test[:60%] will generate the wrong splits because the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples
"""

_URL = "https://www.dropbox.com/s/lc5mje0owl9shms/Guardian.zip?dl=1"


# Using a specific configuration class is optional, you can also use the base class if you don't need
# to add specific attributes.
# here we give an example for three sub-set of the dataset with difference sizes.
class GuardianAuthorshipConfig(datasets.BuilderConfig):
    """BuilderConfig for NewDataset"""

    def __init__(self, train_folder, valid_folder, test_folder, **kwargs):
        """
        Args:
            Train_folder: Topic/genre used for training
            valid_folder:       ~      ~   for validation
            test_folder:        ~      ~   for testing

            **kwargs: keyword arguments forwarded to super.
        """
        super(GuardianAuthorshipConfig, self).__init__(**kwargs)
        self.train_folder = train_folder
        self.valid_folder = valid_folder
        self.test_folder = test_folder


class GuardianAuthorship(datasets.GeneratorBasedBuilder):
    """dataset for same- and cross-topic authorship attribution"""

    config_counter = 0
    BUILDER_CONFIG_CLASS = GuardianAuthorshipConfig
    BUILDER_CONFIGS = [
        # cross-topic
        GuardianAuthorshipConfig(
            name=f"cross_topic_{1}",
            version=datasets.Version(f"{1}.0.0", description=f"The Original DS with the cross-topic scenario no.{1}"),
            train_folder="Politics",
            valid_folder="Society",
            test_folder="UK,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{2}",
            version=datasets.Version(f"{2}.0.0", description=f"The Original DS with the cross-topic scenario no.{2}"),
            train_folder="Politics",
            valid_folder="UK",
            test_folder="Society,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{3}",
            version=datasets.Version(f"{3}.0.0", description=f"The Original DS with the cross-topic scenario no.{3}"),
            train_folder="Politics",
            valid_folder="World",
            test_folder="Society,UK",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{4}",
            version=datasets.Version(f"{4}.0.0", description=f"The Original DS with the cross-topic scenario no.{4}"),
            train_folder="Society",
            valid_folder="Politics",
            test_folder="UK,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{5}",
            version=datasets.Version(f"{5}.0.0", description=f"The Original DS with the cross-topic scenario no.{5}"),
            train_folder="Society",
            valid_folder="UK",
            test_folder="Politics,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{6}",
            version=datasets.Version(f"{6}.0.0", description=f"The Original DS with the cross-topic scenario no.{6}"),
            train_folder="Society",
            valid_folder="World",
            test_folder="Politics,UK",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{7}",
            version=datasets.Version(f"{7}.0.0", description=f"The Original DS with the cross-topic scenario no.{7}"),
            train_folder="UK",
            valid_folder="Politics",
            test_folder="Society,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{8}",
            version=datasets.Version(f"{8}.0.0", description=f"The Original DS with the cross-topic scenario no.{8}"),
            train_folder="UK",
            valid_folder="Society",
            test_folder="Politics,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{9}",
            version=datasets.Version(f"{9}.0.0", description=f"The Original DS with the cross-topic scenario no.{9}"),
            train_folder="UK",
            valid_folder="World",
            test_folder="Politics,Society",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{10}",
            version=datasets.Version(
                f"{10}.0.0", description=f"The Original DS with the cross-topic scenario no.{10}"
            ),
            train_folder="World",
            valid_folder="Politics",
            test_folder="Society,UK",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{11}",
            version=datasets.Version(
                f"{11}.0.0", description=f"The Original DS with the cross-topic scenario no.{11}"
            ),
            train_folder="World",
            valid_folder="Society",
            test_folder="Politics,UK",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_topic_{12}",
            version=datasets.Version(
                f"{12}.0.0", description=f"The Original DS with the cross-topic scenario no.{12}"
            ),
            train_folder="World",
            valid_folder="UK",
            test_folder="Politics,Society",
        ),
        # # cross-genre
        GuardianAuthorshipConfig(
            name=f"cross_genre_{1}",
            version=datasets.Version(f"{1}.0.0", description=f"The Original DS with the cross-genre scenario no.{1}"),
            train_folder="Books",
            valid_folder="Politics",
            test_folder="Society,UK,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_genre_{2}",
            version=datasets.Version(f"{2}.0.0", description=f"The Original DS with the cross-genre scenario no.{2}"),
            train_folder="Books",
            valid_folder="Society",
            test_folder="Politics,UK,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_genre_{3}",
            version=datasets.Version(f"{3}.0.0", description=f"The Original DS with the cross-genre scenario no.{3}"),
            train_folder="Books",
            valid_folder="UK",
            test_folder="Politics,Society,World",
        ),
        GuardianAuthorshipConfig(
            name=f"cross_genre_{4}",
            version=datasets.Version(f"{4}.0.0", description=f"The Original DS with the cross-genre scenario no.{4}"),
            train_folder="Books",
            valid_folder="World",
            test_folder="Politics,Society,UK",
        ),
    ]

    def _info(self):
        # Specifies the datasets.DatasetInfo object
        return datasets.DatasetInfo(
            # This is the description that will appear on the datasets page.
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    # These are the features of your dataset like images, labels ...
                    # There are 13 authors in this dataset
                    "author": datasets.features.ClassLabel(
                        names=[
                            "catherinebennett",
                            "georgemonbiot",
                            "hugoyoung",
                            "jonathanfreedland",
                            "martinkettle",
                            "maryriddell",
                            "nickcohen",
                            "peterpreston",
                            "pollytoynbee",
                            "royhattersley",
                            "simonhoggart",
                            "willhutton",
                            "zoewilliams",
                        ]
                    ),
                    # There are book reviews, and articles on the following four topics
                    "topic": datasets.features.ClassLabel(names=["Politics", "Society", "UK", "World", "Books"]),
                    "article": datasets.Value("string"),
                }
            ),
            # If there's a common (input, target) tuple from the features,
            # specify them here. They'll be used if as_supervised=True in
            # builder.as_dataset.
            supervised_keys=[("article", "author")],
            # Homepage of the dataset for documentation
            homepage="http://www.icsd.aegean.gr/lecturers/stamatatos/papers/JLP2013.pdf",
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        # dl_manager is a datasets.download.DownloadManager that can be used to
        # download and extract URLs
        dl_dir = dl_manager.download_and_extract(_URL)

        # This folder contains the orginal/2013 dataset
        data_dir = os.path.join(dl_dir, "Guardian", "Guardian_original")

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={"data_dir": data_dir, "samples_folders": self.config.train_folder, "split": "train"},
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={"data_dir": data_dir, "samples_folders": self.config.test_folder, "split": "test"},
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={"data_dir": data_dir, "samples_folders": self.config.valid_folder, "split": "valid"},
            ),
        ]

    def _generate_examples(self, data_dir, samples_folders, split):
        """Yields examples."""
        # Yields (key, example) tuples from the dataset

        # Training and validation are on 1 topic/genre, while testing is on multiple topics
        # We convert the sample folders into list (from string)
        if samples_folders.count(",") == 0:
            samples_folders = [samples_folders]
        else:
            samples_folders = samples_folders.split(",")

        # the dataset is structured as:
        # |-Topic1
        # |---author 1
        # |------- article-1
        # |------- article-2
        # |---author 2
        # |------- article-1
        # |------- article-2
        # |-Topic2
        # ...

        for topic in samples_folders:
            full_path = os.path.join(data_dir, topic)

            for author in os.listdir(full_path):

                list_articles = os.listdir(os.path.join(full_path, author))
                if len(list_articles) == 0:
                    # Some authors have no articles on certain topics
                    continue

                for id_, article in enumerate(list_articles):
                    path_2_author = os.path.join(full_path, author)
                    path_2_article = os.path.join(path_2_author, article)

                    with open(path_2_article, "r", encoding="utf8", errors="ignore") as f:
                        art = f.readlines()

                    # The whole article is stored as one line. We access the 1st element of the list
                    # to store it as string, not as a list
                    yield f"{topic}_{author}_{id_}", {
                        "article": art[0],
                        "author": author,
                        "topic": topic,
                    }