Auto-ABSA: Cross-Domain Aspect Detection and Sentiment Analysis Using Auxiliary Sentences

TL;DR

Auto-ABSA addresses cross-domain aspect-based sentiment analysis by integrating SpanEmo-based multi-aspect detection with a domain-transfer sentiment predictor. The Big Model, which combines aspect detection with sentiment classification, enables out-of-domain ABSA using predicted aspects or even NULL aspects, offering robust performance across SemEval and SentiHood datasets. Key findings show that accurate aspect information dramatically improves sentiment predictions, while the integrated Big Model can outperform some ground-truth-aspect baselines, indicating strong generalization. The work advances scalable ABSA by reducing dependence on domain-specific labeled data and providing interpretable per-aspect sentiment reasoning with transformer-based architectures.

Abstract

After transformer is proposed, lots of pre-trained language models have been come up with and sentiment analysis (SA) task has been improved. In this paper, we proposed a method that uses an auxiliary sentence about aspects that the sentence contains to help sentiment prediction. The first is aspect detection, which uses a multi-aspects detection model to predict all aspects that the sentence has. Combining the predicted aspects and the original sentence as Sentiment Analysis (SA) model's input. The second is to do out-of-domain aspect-based sentiment analysis(ABSA), train sentiment classification model with one kind of dataset and validate it with another kind of dataset. Finally, we created two baselines, they use no aspect and all aspects as sentiment classification model's input, respectively. Compare two baselines performance to our method, found that our method really makes sense.

Figures (3)

• Figure 1: The architecture of the SpanEmo model.
• Figure 2: The architecture of the sentiment predictor model.
• Figure 3: The architecture of the Big Model combining SpanEmo and the sentiment predictor for out-of-domain ABSA.