Search

Interactive argument identification is an important research field in dialogical argumentation mining. This task aims to identify the argument pairs with the interactive relationship in the online forum. In this paper, we tackle the task as sentence pair matching. We build our model based on the pre-trained language model (LM) RoBERTa due to its strong ability in modeling semantic information. Based on the peculiarities of the argument texts, we combine the arguments and their corresponding contexts to better identify the interactive relationship of the argument pair. Besides, we adopt data augmentation and vote strategy based on cross-validation to further enhance the performance. Our system rank 5th on track2 of the NLPCC-2021 shared task on Argumentative Text Understanding for AI Debater.

Metaphoric expression is widespread and frequently used to convey emotions. When it comes to metaphor recognition and analysis, there are still not enough samples for these tasks. In this study, we target on recognizing verb metaphors and analyzing their emotions via data augmentation. To this end, we firstly propose a sentence reconstruction method to prune the dependency parsing tree, and thus alleviates the disturbances caused by the noise information. Then, the data augmentation strategies are proposed based on Seq2Seq model and the reconstructed sentence, which generate sufficient candidate samples after an effective quality evaluation. Finally, a proposed model is trained with the extended dataset, and it achieves the recognition and emotion analysis for metaphors. Experiments are conducted on Chinese and English metaphor corpus respectively, and results show that our proposed model has the best performance compared with the baseline methods.