Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic
<p dir="ltr">Hate speech detection in Arabic poses a complex challenge due to the dialectal diversity across the Arab world. Most existing hate speech datasets for Arabic cover only one dialect or one hate speech category. They also lack balance across dialects, topics, and hate/non-...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , , , |
| Published: |
2024
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1864513510345539584 |
|---|---|
| author | Anis Charfi (18697357) |
| author2 | Mabrouka Besghaier (18697360) Raghda Akasheh (18697363) Andria Atalla (18697366) Wajdi Zaghouani (5297402) |
| author2_role | author author author author |
| author_facet | Anis Charfi (18697357) Mabrouka Besghaier (18697360) Raghda Akasheh (18697363) Andria Atalla (18697366) Wajdi Zaghouani (5297402) |
| author_role | author |
| dc.creator.none.fl_str_mv | Anis Charfi (18697357) Mabrouka Besghaier (18697360) Raghda Akasheh (18697363) Andria Atalla (18697366) Wajdi Zaghouani (5297402) |
| dc.date.none.fl_str_mv | 2024-05-30T09:00:00Z |
| dc.identifier.none.fl_str_mv | 10.3389/frai.2024.1391472 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/journal_contribution/Hate_speech_detection_with_ADHAR_a_multi-dialectal_hate_speech_corpus_in_Arabic/26355037 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Information and computing sciences Artificial intelligence Machine learning Language, communication and culture Linguistics natural language processing hate speech Arabic language dialectal Arabic dataset annotation Arabic corpora |
| dc.title.none.fl_str_mv | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| dc.type.none.fl_str_mv | Text Journal contribution info:eu-repo/semantics/publishedVersion text contribution to journal |
| description | <p dir="ltr">Hate speech detection in Arabic poses a complex challenge due to the dialectal diversity across the Arab world. Most existing hate speech datasets for Arabic cover only one dialect or one hate speech category. They also lack balance across dialects, topics, and hate/non-hate classes. In this paper, we address this gap by presenting ADHAR—a comprehensive multi-dialect, multi-category hate speech corpus for Arabic. ADHAR contains 70,369 words and spans four language variants: Modern Standard Arabic (MSA), Egyptian, Levantine, Gulf and Maghrebi. It covers four key hate speech categories: nationality, religion, ethnicity, and race. A major contribution is that ADHAR is carefully curated to maintain balance across dialects, categories, and hate/non-hate classes to enable unbiased dataset evaluation. We describe the systematic data collection methodology, followed by a rigorous annotation process involving multiple annotators per dialect. Extensive qualitative and quantitative analyses demonstrate the quality and usefulness of ADHAR. Our experiments with various classical and deep learning models demonstrate that our dataset enables the development of robust hate speech classifiers for Arabic, achieving accuracy and F1-scores of up to 90% for hate speech detection and up to 92% for category detection. When trained with Arabert, we achieved an accuracy and F1-score of 94% for hate speech detection, as well as 95% for the category detection.</p><h2>Other Information</h2><p dir="ltr">Published in: Frontiers in Artificial Intelligence<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.3389/frai.2024.1391472" target="_blank">https://dx.doi.org/10.3389/frai.2024.1391472</a></p><p dir="ltr">Additional institutions affiliated with: Information Systems Department - CMU-Q</p> |
| eu_rights_str_mv | openAccess |
| id | Manara2_aea39f0e111a108fb48aaca8671a1403 |
| identifier_str_mv | 10.3389/frai.2024.1391472 |
| network_acronym_str | Manara2 |
| network_name_str | Manara2 |
| oai_identifier_str | oai:figshare.com:article/26355037 |
| publishDate | 2024 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in ArabicAnis Charfi (18697357)Mabrouka Besghaier (18697360)Raghda Akasheh (18697363)Andria Atalla (18697366)Wajdi Zaghouani (5297402)Information and computing sciencesArtificial intelligenceMachine learningLanguage, communication and cultureLinguisticsnatural language processinghate speechArabic languagedialectal Arabicdataset annotationArabic corpora<p dir="ltr">Hate speech detection in Arabic poses a complex challenge due to the dialectal diversity across the Arab world. Most existing hate speech datasets for Arabic cover only one dialect or one hate speech category. They also lack balance across dialects, topics, and hate/non-hate classes. In this paper, we address this gap by presenting ADHAR—a comprehensive multi-dialect, multi-category hate speech corpus for Arabic. ADHAR contains 70,369 words and spans four language variants: Modern Standard Arabic (MSA), Egyptian, Levantine, Gulf and Maghrebi. It covers four key hate speech categories: nationality, religion, ethnicity, and race. A major contribution is that ADHAR is carefully curated to maintain balance across dialects, categories, and hate/non-hate classes to enable unbiased dataset evaluation. We describe the systematic data collection methodology, followed by a rigorous annotation process involving multiple annotators per dialect. Extensive qualitative and quantitative analyses demonstrate the quality and usefulness of ADHAR. Our experiments with various classical and deep learning models demonstrate that our dataset enables the development of robust hate speech classifiers for Arabic, achieving accuracy and F1-scores of up to 90% for hate speech detection and up to 92% for category detection. When trained with Arabert, we achieved an accuracy and F1-score of 94% for hate speech detection, as well as 95% for the category detection.</p><h2>Other Information</h2><p dir="ltr">Published in: Frontiers in Artificial Intelligence<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.3389/frai.2024.1391472" target="_blank">https://dx.doi.org/10.3389/frai.2024.1391472</a></p><p dir="ltr">Additional institutions affiliated with: Information Systems Department - CMU-Q</p>2024-05-30T09:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.3389/frai.2024.1391472https://figshare.com/articles/journal_contribution/Hate_speech_detection_with_ADHAR_a_multi-dialectal_hate_speech_corpus_in_Arabic/26355037CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/263550372024-05-30T09:00:00Z |
| spellingShingle | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic Anis Charfi (18697357) Information and computing sciences Artificial intelligence Machine learning Language, communication and culture Linguistics natural language processing hate speech Arabic language dialectal Arabic dataset annotation Arabic corpora |
| status_str | publishedVersion |
| title | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| title_full | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| title_fullStr | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| title_full_unstemmed | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| title_short | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| title_sort | Hate speech detection with ADHAR: a multi-dialectal hate speech corpus in Arabic |
| topic | Information and computing sciences Artificial intelligence Machine learning Language, communication and culture Linguistics natural language processing hate speech Arabic language dialectal Arabic dataset annotation Arabic corpora |