A Cascade Model for Argument Mining in Japanese Political Discussions: the QA Lab-PoliInfo-3 Case Study
This work addresses argument mining in a specific domain (Japanese political discussions), representing an incremental improvement with domain-specific application.
The rVRAIN team tackled the Budget Argument Mining task in Japanese political discussions, achieving best results with a five-class BERT-based cascade model for argument classification and a combination of BERT-based classifier and cosine similarity for information retrieval.
The rVRAIN team tackled the Budget Argument Mining (BAM) task, consisting of a combination of classification and information retrieval sub-tasks. For the argument classification (AC), the team achieved its best performing results with a five-class BERT-based cascade model complemented with some handcrafted rules. The rules were used to determine if the expression was monetary or not. Then, each monetary expression was classified as a premise or as a conclusion in the first level of the cascade model. Finally, each premise was classified into the three premise classes, and each conclusion into the two conclusion classes. For the information retrieval (i.e., relation ID detection or RID), our best results were achieved by a combination of a BERT-based binary classifier, and the cosine similarity of pairs consisting of the monetary expression and budget dense embeddings.