Neural Discourse Structure for Text Categorization

Introduction

Discourse structure represents the organization of a text as a tree. Experimental results show that variants of our model outperform prior work on four out of five tasks considered. In this paper, they investigate the value of discourse structure for text categorization more boardly. In this work, we use a state-of-the-art open-source RST-style discourse parser.

Model

For an internal node $i$, the composition function considers a parent and all of its children. In defining this composition function, we seek for (i.) the contribution of the parent node $e_i$ to be central; and (ii.) the contribution of each child node $e_j$ be determined by its content as well as the discourse relation it holds with the parent.

Result

分享到