Using Machine Learning to Analyze Sudanese Opinions for Political Decision Making

Sentiment Analysis (SA) is the application of text mining techniques for extraction and identification of subjective opinions from textual data. SA has many applications ranging from financial analysis to political decision-making domains. Although a lot of works have been made in SA of English tweets, very limited work focused on Arabic political colloquial tweets. In this paper, we focused on analyzing tweets during Sudanese revolution 2018 and leverage Term Frequency-Inverse Document Frequency and word embedding methods with machine learning classifiers to detect user's sentiments of colloquial Arabic political tweets instead of using modern standard Arabic. Experiments were conducted to evaluate our classifiers and the results showed superiority of the Ensemble learning model with an F-score of 83% compared to other machine learning classifiers.