During Covid-19 pandemic period, people worldwide turned to use social media network to express their opinions and general feelings. Social media platforms like Twitter have become widespread tools for broadcasting and distributing
news and opinions. This paper presents our participation to CERIST Natural Language Processing Challenge, task1.c: Arabic sentiment analysis and fake news detection within covid-19. This complex task is further increased when dealing with dialects that do not have the structure of Modern Standard Arabic (MSA). We introduce an experiment of sentiment analysis of Arabic tweets within covid-19 using machine learning algorithms. The used Arabic dataset was provided by the challenge organizers and it contains 4,128 tweets labeled as Positive, Negative and Neutral for training and 1,034 tweets unlabeled for testing Hadj Ameur & Aliane, 2021. In this experiment the opinions are classified by various machine learning classifiers including Support Vector Machine (SVM), Logistic Regression (LR), Multinomial Naïve Bayes (NB) and K-Nearest Neighbors (KNN). The experimental results indicated that the highest accuracy (94%) was obtained using the Logistic-Regression and SVM among other with training times of 8609s.

Auteurs : Yousra F.G.Elhakeem , Safa Eltayeb,  Mohammed Aldawsari , Omer Salih Dawood Omer

Téléchargement : PDF

Fake news detection has become a major issue in the digital age, with social media playing a major role in its spread. This paper outlines the dataset and methodology used to model Arabic fake news. This paper is about our participation on CERIST Natural Language Processing Challenge. We used the dataset provided for the Task1.c. Arabic sentiment analysis and fake news detection within covid-19. The model used for this task is a simple transformer fake news model based on the Arabic pre-trained language model CAMeL-BERT. This model was utilized in two variants: a fine-tuned model and a Bidirectional long short-term model. The experiment results of this modeling CAMeL-BERT provides the best result by achieving 0.959 F1, thus outperforming all other models variants in detecting fake news.

Auteurs : Mohammed Aldawsari , Omer Salih Dawood Omer  ,Yousra F.G.Elhakeem , Safa Eltayeb

Téléchargement : PDF