Targeted Aspect-Based Sentiment Analysis via Embedding Commonsense Knowledge into an Attentive LSTM

Analyzing people’s opinions and sentiments towards certain aspects is an important task of natural language understanding.In this paper, we propose a novel solution to targeted aspect-based sentiment analysis, which tackles the challengesof both aspect-based sentiment analysis and targeted sentiment analysis by exploiting commonsense knowledge. Weaugment the long short-term memory (LSTM) network with a hierarchical attention mechanism consisting of a target levelattention and a sentence-level attention. Commonsense knowledge of sentiment-related concepts is incorporated intothe end-to-end training of a deep neural network for sentiment classification. In order to tightly integrate the commonsenseknowledge into the recurrent encoder, we propose an extension of LSTM, termed Sentic LSTM. We conduct experimentson two publicly released datasets, which show that the combination of the proposed attention architecture and SenticLSTM can outperform state-of-the-art methods in targeted aspect sentiment tasks.