- Abstract: One of the major goals in automated argumentation mining is to uncover the argument structure present in argumentative text. In order to determine this structure, one must understand how different individual components of the overall argument are linked. General consensus in this field dictates that the argument components form a hierarchy of persuasion, which manifests itself in a tree structure. This work provides the first neural network-based approach to argumentation mining, focusing on extracting links between argument components, with a secondary focus on classifying types of argument components. In order to solve this problem, we propose to use a modification of a Pointer Network architecture. A Pointer Network is appealing for this task for the following reasons: 1) It takes into account the sequential nature of argument components; 2) By construction, it enforces certain properties of the tree structure present in argument relations; 3) The hidden representations can be applied to auxiliary tasks. In order to extend the contribution of the original Pointer Network model, we construct a joint model that simultaneously attempts to learn the type of argument component, as well as continuing to predict links between argument components. The proposed model achieves state-of-the-art results on two separate evaluation corpora. Furthermore, our results show that optimizing for both tasks, as well as adding a fully-connected layer prior to recurrent neural network input, is crucial for high performance.
- TL;DR: We use a modified Pointer Network to predict 1) types of argument components; 2) links between argument components.
- Keywords: Natural language processing
- Conflicts: cs.uml.edu, csail.mit.edu