Document-Level Supervision for Multi-Aspect Sentiment Analysis Without Fine-grained Labels
This work addresses the challenge of expensive fine-grained annotations in ABSA for applications like user reviews, offering a more feasible solution, though it is incremental as it builds on existing topic modeling and VAE methods.
The paper tackles the problem of aspect-based sentiment analysis (ABSA) without fine-grained labels by proposing a VAE-based topic modeling approach that uses document-level supervision, such as overall ratings, and demonstrates significant performance improvements over a state-of-the-art baseline on two benchmark datasets.
Aspect-based sentiment analysis (ABSA) is a widely studied topic, most often trained through supervision from human annotations of opinionated texts. These fine-grained annotations include identifying aspects towards which a user expresses their sentiment, and their associated polarities (aspect-based sentiments). Such fine-grained annotations can be expensive and often infeasible to obtain in real-world settings. There is, however, an abundance of scenarios where user-generated text contains an overall sentiment, such as a rating of 1-5 in user reviews or user-generated feedback, which may be leveraged for this task. In this paper, we propose a VAE-based topic modeling approach that performs ABSA using document-level supervision and without requiring fine-grained labels for either aspects or sentiments. Our approach allows for the detection of multiple aspects in a document, thereby allowing for the possibility of reasoning about how sentiment expressed through multiple aspects comes together to form an observable overall document-level sentiment. We demonstrate results on two benchmark datasets from two different domains, significantly outperforming a state-of-the-art baseline.