Arabic Sentiment Classification using MLP Network Hybrid with Naive Bayes Algorithm

Abstract

Sentiment analysis has recently become one of the growing areas of research related to text mining and natural language processing. Sentiment analysis techniques are increasingly exploited to categorize the opinion text to one or more predefined sentiment classes for the creation and automated maintenance of review-aggregation websites. Most of the current studies related to this topic focus mainly on English texts with very limited resources available for other languages like Arabic. The complexities of Arabic language in morphology, orthography and dialects makes sentiment analysis for Arabic more challenging. In this study, the Naive Bayes algorithm (NB) and Multilayer Perceptron (MLP) network are combined with hybrid system called NB-MLP for Arabic sentiment classification. Five datasets were tested; attraction, hotel, movie, product, and restaurant. The datasets are then classified into positive or negative polarities of sentiment using both standard and combined system. The 10-fold cross validation was employed for splitting the dataset. Over the whole set of experimental data, the results show that the combined system can achieve high classification accuracy and has promising potential application in the Arabic sentiment analysis and opinion mining.