Variational Semi-supervised Aspect-term Sentiment Analysis via Transformer
This work addresses the challenge of fine-grained sentiment analysis for natural language processing applications, offering an incremental improvement by integrating variational inference with transformers to leverage unlabeled data more effectively.
The paper tackles the problem of limited labeled data for aspect-term sentiment analysis by proposing a semi-supervised method using a Variational Autoencoder based on Transformer, which disentangles latent representations to improve sentiment prediction on unlabeled data and enhances classifier performance. Experimental results on SemEval 2014 task 4 show that the method outperforms two general semi-supervised approaches and achieves state-of-the-art performance with four classical classifiers.
Aspect-term sentiment analysis (ATSA) is a longstanding challenge in natural language understanding. It requires fine-grained semantical reasoning about a target entity appeared in the text. As manual annotation over the aspects is laborious and time-consuming, the amount of labeled data is limited for supervised learning. This paper proposes a semi-supervised method for the ATSA problem by using the Variational Autoencoder based on Transformer (VAET), which models the latent distribution via variational inference. By disentangling the latent representation into the aspect-specific sentiment and the lexical context, our method induces the underlying sentiment prediction for the unlabeled data, which then benefits the ATSA classifier. Our method is classifier agnostic, i.e., the classifier is an independent module and various advanced supervised models can be integrated. Experimental results are obtained on the SemEval 2014 task 4 and show that our method is effective with four classical classifiers. The proposed method outperforms two general semisupervised methods and achieves state-of-the-art performance.