Sentiment Classification Via Recurrent Convolutional Neural Networks

Abstract

The state-of-the-art methods used for sentiment classification are primarily based on statistical machine learning, and their performance strongly depends on the quality of the extracted features. The extracted features are often derived from the output of pre-existing natural language processing (NLP) systems, which leads to the propagation of errors in the existing tools and hinders the performance of these systems. In contrast to traditional methods, this paper introduces a recurrent convolutional neural network for text classification that works independently of and without human-designed features. The model applies a recurrent structure to capture as much contextual information as far as possible when learning word representations, which may introduce considerably less noise compared to traditional window-based neural networks. In addition, we also employ a max-pooling layer that automatically judges which words play key roles in sentiment classification to capture the key components in texts. We also conduct experiments on movie review datasets. These experimental results show that the proposed method outperforms current state-of-theart methods.