Deep Attention-Based Review Level Sentiment Analysis for Arabic Reviews

Sentiment analysis is a branch of machine learning that concerns about classifying the polarity for a given text. Recently, it gained a lot of interest because of the availability of huge amount of opinionated data that needs to be analyzed and interpreted. Recently, using deep learning Artificial Neural Network (ANN) architecture has showed significant improvements with high tendency to reveal the underlying semantic meaning in the input text. However, the output of these models could not be explained and the efficiency could not be analyzed because ANN models are considered as a black box and the success of these models comes at the cost of interpretability. The main motivation of this work is to develop Arabic sentiment analysis system that understands review semantics, without using any linguistic resources. Different scenarios and architectures were examined to test the ability of the proposed model to extract salient words out of the input. The results proved the ability of the proposed model to understand a given review by highlighting the most informative words to the class label. In addition, proposed models are supporting visualization option to get intuitive explanation of the output. The effect of applying transfer learning technique on the problem of Arabic sentiment analysis is experimented as well.