Revue de l'Information Scientifique et Technique
Volume 27, Numéro 2, Pages 29-35
2023-11-08
Authors : Elhakeem Yousra F.g . Eltayeb Safa . Aldawsari Mohammed . Salih Dawood Omer Omer .
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. There has been a huge increase in the volume of Arabic posts on tweeter, giving a plenty source of opinions on a variety of topics. This paper aims to introduce an experiment of sentiment analysis task from Arabic tweets within covid-19. This complex task is further increase when dealing with different dialects that do not involve by structure of Modern Standard Arabic (MSA). In this paper, an Arabic Tweet sentiment analysis that uses machine-learning algorithms is proposed. Arabic dataset is collected containing 4,128 tweets labeled as Positive, Negative and Neutral manually for training and 1,034 tweets unlabeled for testing using an API search on Twitter. In this experiment the opinions are classified by various a number of machine learning classifiers including are support vector machine (SVM), logistic regression (LR), MultinomialNB (NB) and KNnearest-neighbours (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.
Sentiment analysis ; support vector machine (SVM) ; logistic regression (LR) ; KNnearest-neighbours (KNN) ; Multinomial (NB)
Aldawsari Mohammed
.
Salih Dawood Omer Omer
.
Yousra F.g Elhakeem
.
Safa Eltayeb
.
pages 62-65.
Bounaama Rabia
.
Abderrahim Mohammed El Amine
.
pages 18-23.
Bennabi Rim Sakina
.
Elberrichi Zakaria
.
pages 26-29.