Aspect Specific Sentiment Analysis using Hierarchical Deep Learning

This paper focuses on the problem of aspect-specific sentiment analysis. The goal here is to not only extract aspects of a product or service, but also to identify specific sentiments being expressed about them. Most existing algorithms address this problem by treating aspect extraction and sentiment analysis as separate phases or by enforcing explicit modeling assumptions on how these two phases should overlap and interact. In this paper, we propose a novel approach based on a hierarchical deep learning framework which overcomes the aforementioned drawbacks. We experiment with various models of semantic compositionality within this framework. Experimental results on real world datasets show that the proposed framework outperforms other state-of-the-art techniques. In addition, we also demonstrate how domain adaptation using word vectors can benefit the task of aspect specific sentiment analyis.