Enhancing Arabic aspect-based sentiment analysis using deep learning models

摘要

Aspect-based sentiment analysis is a special type of sentiment analysis that aims to identify the discussed aspects and their sentiment polarities in a given review. In this paper, two deep learning models are proposed to address essential aspect-based sentiment analysis tasks: aspect-category identification and aspect-sentiment classification. For the first task, an identification model is proposed based on a convolutional neural network and stacked independent long-short term memory. For the second task, a classification model is proposed based on stacked bidirectional independent long-short term memory, a position-weighting mechanism, and multiple attention mechanism layers. The proposed models are evaluated using the Arabic SemEval-2016 dataset for the Hotels domain. Experimental results demonstrate that the proposed models outperform the baseline and other models, where the first model, C-IndyLSTM, achieves an F_1 measure of 58.08%, and the second model, MBRA, achieves an accuracy measure of 87.31%.