GCRE-GPT: A Generative Model for Comparative Relation Extraction
This addresses a specific bottleneck in opinion analysis for natural language processing, but is incremental as it builds on GPT-2.
The paper tackles the problem of directly extracting comparative relations from text, which previous methods could not do, and achieves state-of-the-art accuracy on two datasets.
Given comparative text, comparative relation extraction aims to extract two targets (\eg two cameras) in comparison and the aspect they are compared for (\eg image quality). The extracted comparative relations form the basis of further opinion analysis.Existing solutions formulate this task as a sequence labeling task, to extract targets and aspects. However, they cannot directly extract comparative relation(s) from text. In this paper, we show that comparative relations can be directly extracted with high accuracy, by generative model. Based on GPT-2, we propose a Generation-based Comparative Relation Extractor (GCRE-GPT). Experiment results show that \modelname achieves state-of-the-art accuracy on two datasets.