Hosni Seffih's scientific contributions

Publications (5)

Chapter
Full-text available
The information spread through the Web influences politics, stock markets, public health, people’s reputation and brands. For these reasons, it is crucial to filter out false information. In this paper, we compare different automatic approaches for fake news detection based on statistical text analysis on the vaccination fake news dataset provided...
Conference Paper
Full-text available
The information spread through the Web influences politics, stock markets, public health, people's reputation and brands. For these reasons, it is crucial to filter out false information. In this paper, we compare different automatic approaches for fake news detection based on statistical text analysis on the vaccination fake news dataset provided...
Conference Paper
Full-text available
Automatic identification of Arabic dialects in a text is a difficult task, especially for Maghreb languages and when they are written in Arabic or Latin characters (Arabizi). These texts are characterized by the use of code-switching between the Modern Standard Arabic (MSA) and the Arabic Dialect (AD) in the texts written in Arabic, or between Arab...

Citations

... Another important task addressed in fake news detection is satire detection, with the methods ranging from convolutional neural networks (CNNs) (Guibon et al., 2019) to adversarial training (McHardy et al., 2019) and BERT-based architectures with long-short-term memory (LSTM) (Pandey and Singh, 2022;Liu and Xie, 2021) and CNN (Kaliyar et al., 2021) layers on top. ...
... However, the spread of COVID-19 fake news is a global issue, so it is important to develop machine learning models that can work in multiple languages. Recently, computational approaches have been developed that can deal with fake news in multiple languages [23,[75][76][77], but multilingual biomedical information extraction remains a challenge. MetaMap and scispaCy can be applied only to English text [52,78]. ...
... In order to quickly identify Tunisian texts, we had to organize our crawling based on a keyword search so as to guide the automatic online texts searching and URLs list creating. Thus, we adopted a methodological approach similar to that one outlined in Saadane et al. (2018). With the aim of building a corpus as representative as possible of the linguistic system, we considered useful to identify wide thematic categories that could represent the most common topics of online daily conversations. ...