Academic Phrasebank is an important resource composed of neutral and generic phrases for academic writers. In this paper, we name these neutral and generic phrases reusable phrases, and student writers use them to organize their research articles. Due to the limited size of Academic Phrasebank, it can not meet all the academic writing needs. There are still a large number of reusable phrases in authentic research articles. In order to make up for the deficiency of Academic Phrasebank, we proposed a reusable phrase extraction model based on constituency parsing and dependency parsing to automatically extract reusable phrases from unlabelled research articles. We divided the proposed model into three main components including a reusable words corpus module, a sentence simplification module, and a syntactic parsing module. We created a reusable words corpus of 2129 words to help judge whether a word is neutral and generic, and created two datasets under two scenarios to verify the feasibility of the proposed model.
展开▼
