Encoding dependency representation with convolutional neural network for Target-polarity word collocation extraction

Abstract

Target-polarity word (T-P) collocation extraction is a basic sentiment analysis task, which aims to extract the targets and their modifying polarity words by analyzing the relationships between them. Recent studies rely primarily on syntactic rule matching. However, the syntactic rules are limited and hard matching is always used during the matching procedure that can result in the low recall value. To tackle this problem, we introduce a dependency representation to explore the most useful semantic features behind the syntactic rules and adopt a framework based on a convolutional neural network (CNN) to extract the T-P collocations. The experimental results on four types of product reviews show that our approach can better capture some latent semantic features that the common feature based methods cannot handle, and further significantly outperform other state-of-the-art methods.