Linguistically Regularized LSTMs for Sentiment Classification

TL;DR

This paper introduces linguistically regularized LSTM models for sentence-level sentiment classification that leverage linguistic resources to improve interpretability and performance without requiring expensive phrase-level annotations.

Contribution

It proposes a novel regularization approach that incorporates sentiment lexicons, negation, and intensity words into LSTM models, enhancing sentiment understanding with simple, annotation-efficient methods.

Findings

Effective in capturing sentiment shifts caused by negation and intensity words

Achieves competitive sentiment classification results

Maintains model simplicity while improving linguistic coherence

Abstract

Sentiment understanding has been a long-term goal of AI in the past decades. This paper deals with sentence-level sentiment classification. Though a variety of neural network models have been proposed very recently, however, previous models either depend on expensive phrase-level annotation, whose performance drops substantially when trained with only sentence-level annotation; or do not fully employ linguistic resources (e.g., sentiment lexicons, negation words, intensity words), thus not being able to produce linguistically coherent representations. In this paper, we propose simple models trained with sentence-level annotation, but also attempt to generating linguistically coherent representations by employing regularizers that model the linguistic role of sentiment lexicons, negation words, and intensity words. Results show that our models are effective to capture the sentiment shifting effect of sentiment, negation, and intensity words, while still obtain competitive results without sacrificing the models' simplicity.