Variational Weakly Supervised Sentiment Analysis with Posterior Regularization
This work addresses sentiment analysis for NLP applications by offering an incremental improvement in leveraging weak supervision.
The paper tackled the problem of scarce human annotations in sentiment analysis by proposing a posterior regularization framework for variational weakly supervised methods, resulting in improved performance and more stable predictions with smaller variance.
Sentiment analysis is an important task in natural language processing (NLP). Most of existing state-of-the-art methods are under the supervised learning paradigm. However, human annotations can be scarce. Thus, we should leverage more weak supervision for sentiment analysis. In this paper, we propose a posterior regularization framework for the variational approach to the weakly supervised sentiment analysis to better control the posterior distribution of the label assignment. The intuition behind the posterior regularization is that if extracted opinion words from two documents are semantically similar, the posterior distributions of two documents should be similar. Our experimental results show that the posterior regularization can improve the original variational approach to the weakly supervised sentiment analysis and the performance is more stable with smaller prediction variance.