Suzen Fylke commited on
Commit
3cbc6d5
1 Parent(s): 7533c49

Add sv data and loading script

Browse files
Files changed (3) hide show
  1. README.md +153 -0
  2. kelly.py +100 -0
  3. sv.csv +0 -0
README.md ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ annotations_creators:
3
+ - expert-generated
4
+ language:
5
+ - sv
6
+ language_creators:
7
+ - expert-generated
8
+ license:
9
+ - cc-by-4.0
10
+ multilinguality:
11
+ - monolingual
12
+ pretty_name: kelly
13
+ size_categories:
14
+ - 1K<n<10K
15
+ source_datasets: []
16
+ tags:
17
+ - lexicon
18
+ - swedish
19
+ - CEFR
20
+ task_categories:
21
+ - text-classification
22
+ task_ids:
23
+ - text-scoring
24
+ ---
25
+
26
+ # Dataset Card for Kelly
27
+
28
+ Keywords for Language Learning for Young and adults alike
29
+
30
+ ## Table of Contents
31
+ - [Table of Contents](#table-of-contents)
32
+ - [Dataset Description](#dataset-description)
33
+ - [Dataset Summary](#dataset-summary)
34
+ - [Languages](#languages)
35
+ - [Dataset Structure](#dataset-structure)
36
+ - [Data Instances](#data-instances)
37
+ - [Data Fields](#data-fields)
38
+ - [Data Splits](#data-splits)
39
+ - [Additional Information](#additional-information)
40
+ - [Licensing Information](#licensing-information)
41
+ - [Citation Information](#citation-information)
42
+ - [Contributions](#contributions)
43
+
44
+ ## Dataset Description
45
+
46
+ - **Homepage:** https://spraakbanken.gu.se/en/resources/kelly
47
+ - **Paper:** https://link.springer.com/article/10.1007/s10579-013-9251-2
48
+
49
+ ### Dataset Summary
50
+
51
+ The Swedish Kelly list is a freely available frequency-based vocabulary list
52
+ that comprises general-purpose language of modern Swedish. The list was
53
+ generated from a large web-acquired corpus (SweWAC) of 114 million words
54
+ dating from the 2010s. It is adapted to the needs of language learners and
55
+ contains 8,425 most frequent lemmas that cover 80% of SweWAC.
56
+
57
+ Please refer to the article [Corpus-based approaches for the creation of a frequency
58
+ based vocabulary list in the EU project KELLY – issues on reliability, validity and
59
+ coverage](https://gup.ub.gu.se/publication/148533?lang=en) for information about how
60
+ the dataset was created.
61
+
62
+ ### Languages
63
+
64
+ Swedish (sv-SE)
65
+
66
+ ## Dataset Structure
67
+
68
+ ### Data Instances
69
+
70
+ Here is a sample of the data:
71
+
72
+ ```python
73
+ {
74
+ 'id': '190',
75
+ 'raw_frequency': 117835.0,
76
+ 'relative_frequency': 1033.61,
77
+ 'cefr_level': 'A1',
78
+ 'source': 'SweWaC',
79
+ 'marker': 'en',
80
+ 'lemma': 'dag',
81
+ 'pos': 'noun-en',
82
+ 'examples': 'e.g. god dag'
83
+ }
84
+ ```
85
+
86
+ This can be understood as:
87
+
88
+ > The common noun "dag" ("day") has a rank of 190 in the list. It was used 117,835
89
+ times in SweWaC, meaning it occured 1033.61 times per million words. This word
90
+ is among the most important vocabulary words for Swedish language learners and
91
+ should be learned at the A1 CEFR level. An example usage of this word is the
92
+ phrase "god dag" ("good day").
93
+
94
+
95
+ ### Data Fields
96
+
97
+ - `id`: The row number for the data entry, starting at 1. Generally corresponds
98
+ to the rank of the word.
99
+ - `raw_frequency`: The raw frequency of the word or -1.0 if not available.
100
+ - `relative_frequency`: The relative frequency of the word measured in
101
+ number of occurences per million words or -1.0 if not available.
102
+ - `cefr_level`: The CEFR level (A1, A2, B1, B2, C1, C2) of the word.
103
+ - `source`: Whether the word came from SweWAC, translation lists (T2), or
104
+ was manually added (manual).
105
+ - `marker`: The grammatical marker of the word, if any, such as an article or
106
+ infinitive marker.
107
+ - `lemma`: The lemma of the word, sometimes provided with its spelling or
108
+ stylistic variants.
109
+ - `pos`: The word's part-of-speech.
110
+ - `examples`: Usage examples and comments. Only available for some of the words.
111
+
112
+ Manual entries were prepended to the list, giving them a higher rank than they
113
+ might otherwise have had. For example, the manual entry "Göteborg ("Gothenberg")
114
+ has a rank of 20, while the first non-manual entry "och" ("and") has a rank of
115
+ 87. However, a conjunction and common stopword is far more likely to occur than
116
+ the name of a city.
117
+
118
+ Additionally, manually entries hav a recoreded relative frequency of `1000000`.
119
+
120
+ ### Data Splits
121
+
122
+ There is a single split, `train`.
123
+
124
+ ## Additional Information
125
+
126
+ ### Licensing Information
127
+
128
+ [CC BY 4.0](https://creativecommons.org/licenses/by/4.0)
129
+
130
+ ### Citation Information
131
+
132
+ Please cite the authors if you use this dataset in your work:
133
+
134
+ ```bibtex
135
+ @article{Kilgarriff2013,
136
+ doi = {10.1007/s10579-013-9251-2},
137
+ url = {https://doi.org/10.1007/s10579-013-9251-2},
138
+ year = {2013},
139
+ month = sep,
140
+ publisher = {Springer Science and Business Media {LLC}},
141
+ volume = {48},
142
+ number = {1},
143
+ pages = {121--163},
144
+ author = {Adam Kilgarriff and Frieda Charalabopoulou and Maria Gavrilidou and Janne Bondi Johannessen and Saussan Khalil and Sofie Johansson Kokkinakis and Robert Lew and Serge Sharoff and Ravikiran Vadlapudi and Elena Volodina},
145
+ title = {Corpus-based vocabulary lists for language learners for nine languages},
146
+ journal = {Language Resources and Evaluation}
147
+ }
148
+ ```
149
+
150
+ ### Contributions
151
+
152
+ Thanks to [@spraakbanken](https://github.com/spraakbanken) for creating this dataset
153
+ and to [@codesue](https://github.com/codesue) for adding it.
kelly.py ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ The dataset loading script for the codesue/kelly dataset.
3
+ """
4
+
5
+ import csv
6
+
7
+ import datasets
8
+
9
+ _CITATION = """\
10
+ @article{Kilgarriff2013,
11
+ doi = {10.1007/s10579-013-9251-2},
12
+ url = {https://doi.org/10.1007/s10579-013-9251-2},
13
+ year = {2013},
14
+ month = sep,
15
+ publisher = {Springer Science and Business Media {LLC}},
16
+ volume = {48},
17
+ number = {1},
18
+ pages = {121--163},
19
+ author = {Adam Kilgarriff and Frieda Charalabopoulou and Maria Gavrilidou and Janne Bondi Johannessen and Saussan Khalil and Sofie Johansson Kokkinakis and Robert Lew and Serge Sharoff and Ravikiran Vadlapudi and Elena Volodina},
20
+ title = {Corpus-based vocabulary lists for language learners for nine languages},
21
+ journal = {Language Resources and Evaluation}
22
+ }
23
+ """
24
+
25
+ _DESCRIPTION = """\
26
+ The Swedish Kelly list is a freely available frequency-based vocabulary list \
27
+ that comprises general-purpose language of modern Swedish. The list was \
28
+ generated from a large web-acquired corpus (SweWAC) of 114 million words \
29
+ dating from the 2010s. It is adapted to the needs of language learners \
30
+ and contains 8,425 most frequent lemmas that cover 80% of SweWAC.
31
+ """
32
+
33
+ _HOMEPAGE = "https://spraakbanken.gu.se/en/resources/kelly"
34
+
35
+ _LICENSE = "CC BY 4.0"
36
+
37
+ _URLS = {
38
+ "csv": "sv.csv",
39
+ }
40
+
41
+
42
+ class Kelly(datasets.GeneratorBasedBuilder):
43
+ """Kelly: Keywords for Language Learning for Young and adults alike"""
44
+
45
+ VERSION = datasets.Version("1.0.0")
46
+
47
+ def _info(self):
48
+ features = datasets.Features(
49
+ {
50
+ "id": datasets.Value("string"),
51
+ "raw_frequency": datasets.Value("float64"),
52
+ "relative_frequency": datasets.Value("float64"),
53
+ "cefr_level": datasets.Value("string"),
54
+ "source": datasets.Value("string"),
55
+ "marker": datasets.Value("string"),
56
+ "lemma": datasets.Value("string"),
57
+ "class": datasets.Value("string"),
58
+ "examples": datasets.Value("string"),
59
+ }
60
+ )
61
+
62
+ return datasets.DatasetInfo(
63
+ description=_DESCRIPTION,
64
+ features=features,
65
+ homepage=_HOMEPAGE,
66
+ license=_LICENSE,
67
+ citation=_CITATION,
68
+ )
69
+
70
+ def _split_generators(self, dl_manager):
71
+ data_path = dl_manager.download_and_extract(_URLS["csv"])
72
+ return [
73
+ datasets.SplitGenerator(
74
+ name=datasets.Split.TRAIN,
75
+ gen_kwargs={
76
+ "filepath": data_path,
77
+ },
78
+ ),
79
+ ]
80
+
81
+ def _generate_examples(self, filepath):
82
+ """Generate text2log dataset examples."""
83
+ with open(filepath, encoding="utf-8") as csv_file:
84
+ csv_reader = csv.reader(
85
+ csv_file, quotechar='"', delimiter=",", quoting=csv.QUOTE_ALL, skipinitialspace=True
86
+ )
87
+ next(csv_reader)
88
+ for key, row in enumerate(csv_reader):
89
+ a, b, c, d, e, f, g, h, i = row
90
+ yield key, {
91
+ "id": a,
92
+ "raw_frequency": b or "-1",
93
+ "relative_frequency": c or "-1",
94
+ "cefr_level": d,
95
+ "source": e,
96
+ "marker": f,
97
+ "lemma": g,
98
+ "class": h,
99
+ "examples": i,
100
+ }
sv.csv ADDED
The diff for this file is too large to render. See raw diff