A Joint Framework for Argumentative Text Analysis Incorporating Domain Knowledge
This work addresses the challenge of integrating multiple sub-tasks in argumentation mining for researchers and practitioners in natural language processing, representing an incremental improvement over existing methods.
The paper tackles the problem of argumentation mining by proposing a joint framework that incorporates logical relations between sub-tasks, such as component type and relation classification, to improve argumentation structure generation. The results show that the model outperforms separate baseline models significantly on two public corpora and has advantages over a state-of-the-art joint model for component-related tasks.
For argumentation mining, there are several sub-tasks such as argumentation component type classification, relation classification. Existing research tends to solve such sub-tasks separately, but ignore the close relation between them. In this paper, we present a joint framework incorporating logical relation between sub-tasks to improve the performance of argumentation structure generation. We design an objective function to combine the predictions from individual models for each sub-task and solve the problem with some constraints constructed from background knowledge. We evaluate our proposed model on two public corpora and the experiment results show that our model can outperform the baseline that uses a separate model significantly for each sub-task. Our model also shows advantages on component-related sub-tasks compared to a state-of-the-art joint model based on the evidence graph.