Abstract
Sentiment analysis is the process of analyzing people's opinion and feelings toward individual, entities, issues, items or topics. Within the last couple of years, this field has gained increasing research interest varying from simple linear models to more complex neural networks models. Several challenges including the grammar, structures of the language and Morphology which one-word lead to many important meanings. The varieties of dialects and the lack of the appropriate corpora limit the use of sentiment analysis on Arabic content. Throughout this paper, we aim at providing a review on the utilization of the deep learning (DL) approach to analyze sentiments expressed in the Arabic text by using Multi-layer perceptron (MLP) model. The results show that the MLP model shows highly effective performance in sentiment analysis for both Arabic and English languages