Menel Mahamdi


2025

pdf bib
Beyond Dataset Creation: Critical View of Annotation Variation and Bias Probing of a Dataset for Online Radical Content Detection
Arij Riabi | Virginie Mouilleron | Menel Mahamdi | Wissam Antoun | Djamé Seddah
Proceedings of the 31st International Conference on Computational Linguistics

The proliferation of radical content on online platforms poses significant risks, including inciting violence and spreading extremist ideologies. Despite ongoing research, existing datasets and models often fail to address the complexities of multilingual and diverse data. To bridge this gap, we introduce a publicly available multilingual dataset annotated with radicalization levels, calls for action, and named entities in English, French, and Arabic. This dataset is pseudonymized to protect individual privacy while preserving contextual information. Beyond presenting our freely available dataset, we analyze the annotation process, highlighting biases and disagreements among annotators and their implications for model performance. Additionally, we use synthetic data to investigate the influence of socio-demographic traits on annotation patterns and model predictions. Our work offers a comprehensive examination of the challenges and opportunities in building robust datasets for radical content detection, emphasizing the importance of fairness and transparency in model development. The Counter dataset is available at https://gitlab.inria.fr/ariabi/counter-dataset-public.

pdf bib
Peut-on retrouver votre âge à partir de la transcription de votre parole ?
Vanessa Gaudray Bouju | Menel Mahamdi | Iris Eshkol-Taravella | Angèle Barbedette
Actes des 32ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN), volume 1 : articles scientifiques originaux

L’identification et la classification des groupes sociaux à partir du langage constitue une préoccupation sociolinguistique majeure. Dans cet article, nous présentons une recherche de classification des locuteurs basée sur leur âge. Pour ce faire, nous exploitons un corpus de données du français oral, où chaque locuteur est associé à des métadonnées, dont son âge au moment de l’enregistrement. Notre objectif est de développer des méthodes d’apprentissage automatique capables de prédire la tranche d’âge d’un locuteur à partir de son discours transcrit de l’oral, allant de l’apprentissage supervisé à l’ingénierie de prompts sur des grands modèles de langage. Cette tâche n’est pas seulement un défi technique, elle soulève également des questions fondamentales sur la nature de la variation linguistique et sur les liens entre le langage et la société. En effet, en identifiant les corrélations entre certains traits linguistiques et l’âge, notre projet contribue à enrichir notre compréhension des mécanismes sous-jacents à la variation du langage et à ses implications dans la construction de l’identité sociale. Son autre apport est de questionner les traits linguistiques classiquement imputés à une tranche d’âge afin de montrer leurs limites.

2024

pdf bib
Cloaked Classifiers: Pseudonymization Strategies on Sensitive Classification Tasks
Arij Riabi | Menel Mahamdi | Virginie Mouilleron | Djamé Seddah
Proceedings of the Fifth Workshop on Privacy in Natural Language Processing

Protecting privacy is essential when sharing data, particularly in the case of an online radicalization dataset that may contain personal information. In this paper, we explore the balance between preserving data usefulness and ensuring robust privacy safeguards, since regulations like the European GDPR shape how personal information must be handled. We share our method for manually pseudonymizing a multilingual radicalization dataset, ensuring performance comparable to the original data. Furthermore, we highlight the importance of establishing comprehensive guidelines for processing sensitive NLP data by sharing our complete pseudonymization process, our guidelines, the challenges we encountered as well as the resulting dataset.

2023

pdf bib
Enriching the NArabizi Treebank: A Multifaceted Approach to Supporting an Under-Resourced Language
Arij Riabi | Menel Mahamdi | Djamé Seddah
Proceedings of the 17th Linguistic Annotation Workshop (LAW-XVII)

In this paper we address the scarcity of annotated data for NArabizi, a Romanized form of North African Arabic used mostly on social media, which poses challenges for Natural Language Processing (NLP). We introduce an enriched version of NArabizi Treebank (Seddah et al., 2020) with three main contributions: the addition of two novel annotation layers (named entity recognition and offensive language detection) and a re-annotation of the tokenization, morpho-syntactic and syntactic layers that ensure annotation consistency. Our experimental results, using different tokenization schemes, showcase the value of our contributions and highlight the impact of working with non-gold tokenization for NER and dependency parsing. To facilitate future research, we make these annotations publicly available. Our enhanced NArabizi Treebank paves the way for creating sophisticated language models and NLP tools for this under-represented language.