Hate speech detection in tweets / Islem Abdelhakim Maboud  

Public ISBD Titre : Hate speech detection in tweets Type de document : texte imprimé Auteurs : Islem Abdelhakim Maboud, Auteur ; yasmine Agoun, Auteur ; Refoufi, Allaoua, Directeur de thèse Année de publication : 2022 Importance : 1 vol (59 f .) Format : 29cm Langues : Français (fre) Catégories : Thèses & Mémoires:Informatique Mots-clés : hate speech machine learning Index. décimale : 004 Informatique Résumé : In recent years, Hate speech online is constantly increasing on different social media platforms by using aggressive, violent or offensive language to a specific individual or group , which has become a major issue . Due to the size of social media and unfiltered feed of messages posted in social media that can contain hate speech targeted, we are interested in how to reduce hate on social media. In an effort to identify solutions for the problem of hate speech in social media, we propose an approach to automatically classify tweets on Twitter into two classes: hate speech and neutral speech. Using the Twitter dataset with deep learning using BERT based on transformers to learn contextual relations between words,we achieved an accuracy of 95.9%.and a recall of 95.9%.and F1 of 95.9%.. We also developed a React web application using tensorflow-serving serves as a real time application of our model ,the user can enter a tweet given in the text box and know if it’s a hateful expression or not. Côte titre : MAI/0614 En ligne : https://drive.google.com/file/d/1ffZp67082LaQmk11oXyonSCIs6qu82U8/view?usp=share [...] Format de la ressource électronique : pdf Hate speech detection in tweets [texte imprimé] / Islem Abdelhakim Maboud, Auteur ; yasmine Agoun, Auteur ; Refoufi, Allaoua, Directeur de thèse . - 2022 . - 1 vol (59 f .) ; 29cm. Langues : Français (fre) Catégories : Thèses & Mémoires:Informatique Mots-clés : hate speech machine learning Index. décimale : 004 Informatique Résumé : In recent years, Hate speech online is constantly increasing on different social media platforms by using aggressive, violent or offensive language to a specific individual or group , which has become a major issue . Due to the size of social media and unfiltered feed of messages posted in social media that can contain hate speech targeted, we are interested in how to reduce hate on social media. In an effort to identify solutions for the problem of hate speech in social media, we propose an approach to automatically classify tweets on Twitter into two classes: hate speech and neutral speech. Using the Twitter dataset with deep learning using BERT based on transformers to learn contextual relations between words,we achieved an accuracy of 95.9%.and a recall of 95.9%.and F1 of 95.9%.. We also developed a React web application using tensorflow-serving serves as a real time application of our model ,the user can enter a tweet given in the text box and know if it’s a hateful expression or not. Côte titre : MAI/0614 En ligne : https://drive.google.com/file/d/1ffZp67082LaQmk11oXyonSCIs6qu82U8/view?usp=share [...] Format de la ressource électronique : pdf

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MAI/0614	MAI/0614	Mémoire	Bibliothéque des sciences	Anglais	Disponible Disponible

---

1 (1 - 1 / 1)