smart.lab.tunisia@gmail.com +216 71 588 443
Connexion

Natural langage processing

Description

Aucune description disponible pour cet axe de recherche.

Membres

Chayma Sakrani Chayma Sakrani

Publications

  • 2023
    Chayma sakrani, Boutheina Jlifi

    Towards a soft three-level voting model (Soft T-LVM) for fake news detection

    Journal of Intelligent Information Systems, 61(1), 249-269., 2023

    Résumé

    Fake news has a worldwide impact and the potential to change political scenarios and human behavior, especially in a critical time like the COVID-19 pandemic. This work suggests a Soft Three-Level Voting Model (Soft T-LVM) for automatically classifying COVID-19 fake news. We train different individual machine learning algorithms and different ensemble methods in order to overcome the weakness of individual models. This novel model is based on the soft-voting technique to calculate the class with the majority of votes and to choose the classifiers to merge and apply at every level. We use the Grid search method to tune the hyper-parameters during the process of classification and voting. The experimental evaluation confirms that our proposed model approach has superior performance compared to the other classifiers.

    Chayma sakrani, Boutheina Jlifi

    Towards a soft three-level voting model (Soft T-LVM) for fake news detection

    Journal of Intelligent Information Systems, 61(1), 249-269., 2023

    Résumé

    Fake news has a worldwide impact and the potential to change political scenarios and human behavior, especially in a critical time like the COVID-19 pandemic. This work suggests a Soft Three-Level Voting Model (Soft T-LVM) for automatically classifying COVID-19 fake news. We train different individual machine learning algorithms and different ensemble methods in order to overcome the weakness of individual models. This novel model is based on the soft-voting technique to calculate the class with the majority of votes and to choose the classifiers to merge and apply at every level. We use the Grid search method to tune the hyper-parameters during the process of classification and voting. The experimental evaluation confirms that our proposed model approach has superior performance compared to the other classifiers.

  • Jihene Sassi, Ines Thabet, Khaled Ghedira

    A Framework to Support Tunisian Tweets Analysis for Crisis Management

    ISCRAM-med 2017, 2017

    Résumé

    The increasing crisis frequency and the growing impact of their damages require efficient crisis management processes in order to manage crisis effectively and reduce losses. In such context, the need of accurate and updated information about crises is extremely important. In recent years, crisis information has frequently been provided by social media platforms such as Twitter, Facebook, Flickr, etc. In fact, considering the huge amount of shared information, their precision and their real time characteristic, organizations are moving towards the development of crisis management applications that include information provided by social media platforms. Following this view, the main purpose of our work is to propose a framework for Tunisian tweets extraction and analysis. More precisely, we provide an architecture that includes necessary components and tools for Tunisian dialect treatment. The proposed architecture is an extension on the existing AIDR platform. In addition, we specify the functioning of the proposed architecture to enable firstly terms transliteration from Arabic to Latin alphabet, secondly their normalization and finally their translation in order to be treated by existing social media analysis platform.