CONJR: Conjunctive Sentence Splitter without Parsing

Anonymous

Abstract: In this paper, we observe and address the challenges of splitting conjunctive sentences around each group of conjuncts. Most existing methods rely on parsers to identify the conjuncts in a sentence and detect the coordination boundaries. However, state-of-the-art syntactic parsers are slow and suffer from errors, especially for long and complicated sentences. In order to better solve the problems, we formulate coordination boundary detection as a sequence tagging task and propose a specialized model CONJR without using syntactic parsers. We introduce both semantic and syntactic features and a specially designed attention mechanism to capture the symmetry among the potential conjuncts. The experimental results on datasets from various domains demonstrate the effectiveness of our proposed methods.