Classification of Arabic Social Media Data

摘要

Arabic sentiment analysis is a process for classifying unseen Arabic text to their predefined classes (i.e. Positive, Negative and Neutral). This paper investigates Decision tree, Naive Bayesian (NB), K Nearest Neighbor (KNN) and Support Vector Machines (SVMs) on Arabic Twitter corpus. Our Arabic Twitter corpus is collected from twitter ARCHIVISIT. Our experimental results conducted based on the most well-known sentiment analysis evaluation measures. The Experimental results against Arabic Twitter corpus indicate that the SVMs learning method outperformed the KNN, NB and Decision tree learning methods with regards to Recall, Precision and F1 measures.