Hugging Face's logo

Datasets:
ctoraman
/
large-scale-hate-speech

Tasks:
Text Classification
Languages:
English
Turkish
Size:
100K<n<1M
Tags:
hate-speech
hatespeech
hate-speech-detection
hatespeechdetection
License:
Dataset card Data Studio Files
xet
Community
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code:   DatasetGenerationCastError
Exception:    DatasetGenerationCastError
Message:      An error occurred while generating the dataset

All the data files must have the same columns, but at some point there are 5 missing columns ({'Label_2', 'Label_3', 'Label_4', 'Label_1', 'Label_5'})

This happened while the csv dataset builder was generating data using

hf://datasets/ctoraman/large-scale-hate-speech/toraman22_hate_speech_dataset_v2.csv (at revision 6b5403b383110f85519b4afef56dbd59c864dcf5)

Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 2011, in _prepare_split_single
                  writer.write_table(table)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/arrow_writer.py", line 585, in write_table
                  pa_table = table_cast(pa_table, self._schema)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2302, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/table.py", line 2256, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              TweetID: int64
              LangID: int64
              TopicID: int64
              HateLabel: int64
              -- schema metadata --
              pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 709
              to
              {'TweetID': Value(dtype='int64', id=None), 'LangID': Value(dtype='int64', id=None), 'TopicID': Value(dtype='int64', id=None), 'Label_1': Value(dtype='int64', id=None), 'Label_2': Value(dtype='int64', id=None), 'Label_3': Value(dtype='int64', id=None), 'Label_4': Value(dtype='int64', id=None), 'Label_5': Value(dtype='int64', id=None), 'HateLabel': Value(dtype='int64', id=None)}
              because column names don't match
              
              During handling of the above exception, another exception occurred:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1321, in compute_config_parquet_and_info_response
                  parquet_operations = convert_to_parquet(builder)
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 935, in convert_to_parquet
                  builder.download_and_prepare(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1027, in download_and_prepare
                  self._download_and_prepare(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1122, in _download_and_prepare
                  self._prepare_split(split_generator, **prepare_split_kwargs)
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 1882, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                File "/src/services/worker/.venv/lib/python3.9/site-packages/datasets/builder.py", line 2013, in _prepare_split_single
                  raise DatasetGenerationCastError.from_cast_error(
              datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
              
              All the data files must have the same columns, but at some point there are 5 missing columns ({'Label_2', 'Label_3', 'Label_4', 'Label_1', 'Label_5'})
              
              This happened while the csv dataset builder was generating data using
              
              hf://datasets/ctoraman/large-scale-hate-speech/toraman22_hate_speech_dataset_v2.csv (at revision 6b5403b383110f85519b4afef56dbd59c864dcf5)
              
              Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

TweetID
int64
LangID
int64
TopicID
int64
Label_1
int64
Label_2
int64
Label_3
int64
Label_4
int64
Label_5
int64
HateLabel
int64
1,310,773,615,709,827,000
0
0
2
2
2
1
2
2
1,233,309,088,181,899,300
0
0
2
2
2
2
2
2
1,303,626,209,788,678,100
0
0
2
2
2
0
2
2
1,346,887,994,721,054,700
0
0
2
0
2
0
2
2
1,323,499,874,864,763,000
0
0
2
0
2
0
2
2
1,317,405,083,714,412,500
0
0
1
2
2
2
1
2
1,347,062,948,045,402,000
0
0
2
2
2
2
2
2
1,314,962,835,903,000,600
0
0
0
2
1
2
2
2
1,262,181,976,796,209,200
0
0
0
2
2
1
2
2
1,343,677,953,926,983,700
0
0
2
1
2
2
2
2
1,321,394,510,740,529,200
0
0
2
1
2
2
2
2
1,265,497,874,521,108,500
0
0
2
2
2
1
0
2
1,337,083,746,420,404,200
0
0
1
2
2
2
0
2
1,365,866,579,414,700,000
0
0
2
2
2
2
2
2
1,256,298,017,004,675,000
0
0
0
2
1
2
2
2
1,233,766,782,751,645,700
0
0
1
1
2
2
2
2
1,329,333,713,914,765,300
0
0
2
1
2
1
2
2
1,317,573,195,046,813,700
0
0
2
2
2
2
1
2
1,347,569,983,619,002,400
0
0
2
0
2
2
0
2
1,344,020,813,960,388,600
0
0
2
2
2
2
2
2
1,345,741,148,934,045,700
0
0
2
0
2
2
2
2
1,369,404,382,463,340,500
0
0
2
0
2
2
2
2
1,345,409,648,199,938,000
0
0
1
1
2
2
2
2
1,343,890,764,703,670,300
0
0
2
2
2
1
1
2
1,363,010,463,345,246,200
0
0
2
2
2
2
2
2
1,290,875,253,221,138,400
0
0
2
2
2
0
1
2
1,245,806,844,594,884,600
0
0
2
2
1
1
2
2
1,371,847,471,312,343,000
0
0
2
2
1
2
2
2
1,342,411,660,523,401,200
0
0
1
2
2
2
1
2
1,303,378,229,009,539,000
0
0
2
2
2
2
2
2
1,233,883,721,746,571,300
0
0
2
1
2
2
1
2
1,253,389,343,907,033,000
0
0
2
2
2
1
2
2
1,259,147,100,652,212,200
0
0
2
1
1
2
2
2
1,316,066,807,745,310,700
0
0
0
2
2
2
2
2
1,233,878,731,736,916,000
0
0
2
2
1
2
2
2
1,233,844,908,022,280,200
0
0
2
2
1
2
1
2
1,233,726,734,207,537,200
0
0
2
2
1
2
1
2
1,361,724,190,525,689,900
0
0
2
2
2
2
2
2
1,342,001,801,948,782,600
0
0
2
1
2
2
2
2
1,274,461,355,626,696,700
0
0
2
2
2
2
2
2
1,278,061,068,720,509,000
0
0
2
0
2
2
1
2
1,307,325,966,956,535,800
0
0
2
1
2
1
2
2
1,263,276,598,704,443,400
0
0
2
1
1
2
2
2
1,255,112,491,086,106,600
0
0
1
2
2
1
2
2
1,270,836,385,507,938,300
0
0
1
1
2
2
2
2
1,344,294,488,144,048,000
0
0
2
2
2
1
2
2
1,313,487,690,172,686,300
0
0
1
2
2
2
2
2
1,373,350,200,119,529,500
0
0
1
2
2
0
2
2
1,357,666,605,384,237,000
0
0
1
0
2
2
2
2
1,357,056,075,070,070,800
0
0
2
2
2
2
2
2
1,233,857,313,909,485,600
0
0
2
2
2
2
2
2
1,342,947,370,158,461,000
0
0
2
2
2
2
2
2
1,233,828,864,213,835,800
0
0
2
2
2
2
2
2
1,270,863,431,436,992,500
0
0
2
2
2
1
2
2
1,371,564,297,692,790,800
0
0
2
2
2
2
2
2
1,342,241,048,597,323,800
0
0
2
2
2
2
1
2
1,233,823,307,297,890,300
0
0
2
0
2
2
2
2
1,323,391,547,803,652,000
0
0
1
2
2
2
2
2
1,250,118,905,508,610,000
0
0
1
1
2
2
2
2
1,346,923,491,262,685,200
0
0
2
2
2
2
2
2
1,263,505,172,032,094,200
0
0
1
2
1
2
2
2
1,343,582,048,032,870,400
0
0
2
2
2
2
2
2
1,320,255,468,036,591,600
0
0
2
2
0
2
2
2
1,317,340,416,795,136,000
0
0
2
2
2
0
2
2
1,272,438,899,088,994,300
0
0
2
2
2
2
2
2
1,364,278,673,239,658,500
0
0
2
2
2
2
2
2
1,346,839,200,423,948,300
0
0
0
2
0
2
2
2
1,317,388,764,541,575,200
0
0
2
0
1
2
2
2
1,325,094,688,093,827,000
0
0
2
2
2
2
1
2
1,329,853,986,585,927,700
0
0
2
1
2
0
2
2
1,233,873,706,365,128,700
0
0
1
1
2
2
2
2
1,371,828,953,267,253,200
0
0
2
2
2
2
0
2
1,317,462,022,838,194,200
0
0
2
2
2
2
2
2
1,267,248,190,132,236,300
0
0
1
2
2
1
2
2
1,341,456,003,745,902,600
0
0
2
2
2
1
2
2
1,260,904,868,732,113,000
0
0
2
1
1
2
2
2
1,341,814,560,630,710,300
0
0
2
2
2
2
2
2
1,233,430,228,388,765,700
0
0
2
2
2
2
2
2
1,335,625,994,879,824,000
0
0
2
2
2
2
1
2
1,254,670,071,873,638,400
0
0
2
1
2
1
2
2
1,323,948,358,336,192,500
0
0
2
1
2
2
2
2
1,305,945,472,796,889,000
0
0
1
2
2
2
2
2
1,233,309,872,831,291,400
0
0
2
2
2
2
2
2
1,233,171,120,297,271,300
0
0
2
2
0
2
1
2
1,328,441,350,992,367,600
0
0
2
2
2
0
2
2
1,233,688,952,558,825,500
0
0
2
2
2
0
0
2
1,333,891,203,516,014,600
0
0
2
1
0
2
2
2
1,290,762,732,678,066,200
0
0
2
2
2
1
2
2
1,317,405,602,549,813,200
0
0
2
2
2
2
1
2
1,323,493,632,213,684,200
0
0
2
2
1
2
1
2
1,341,655,595,984,810,000
0
0
2
2
2
2
2
2
1,355,642,779,347,677,200
0
0
2
2
2
2
2
2
1,315,630,293,462,679,600
0
0
2
2
2
2
2
2
1,270,726,083,197,808,600
0
0
2
1
2
2
2
2
1,366,081,200,926,957,600
0
0
1
0
2
2
2
2
1,260,530,584,469,848,000
0
0
1
2
1
2
2
2
1,342,153,625,955,668,000
0
0
2
2
2
2
2
2
1,310,253,261,203,689,500
0
0
0
2
2
0
2
2
1,262,155,143,908,376,600
0
0
2
2
1
1
2
2
1,329,533,363,435,278,300
0
0
2
2
2
1
0
2
End of preview.

This repository contains the utilized dataset in the LREC 2022 paper "Large-Scale Hate Speech Detection with Cross-Domain Transfer". This study mainly focuses hate speech detection in Turkish and English. In addition, domain transfer success between hate domains is also examined.

There are two dataset versions.

Dataset v1: The original dataset that includes 100,000 tweets per English and Turkish, published in LREC 2022. The annotations with more than 60% agreement are included.

Dataset v2: A more reliable dataset version that includes 68,597 tweets for English and 60,310 for Turkish. The annotations with more than 80% agreement are included.

NOTE: We publish all dataset contents in the Github repo. We also publish ~20k image files associated with the tweets. https://github.com/metunlp/hate-speech

Downloads last month
38