作者: Lingpeng Kong , Alexander M. Rush , Noah A. Smith
DOI: 10.3115/V1/N15-1080
关键词:
摘要: We present a new algorithm for transforming dependency parse trees into phrase-structure trees. cast the problem as structured prediction and learn statistical model. Our is faster than traditional phrasestructure parsing achieves 90.4% English accuracy 82.4% Chinese accuracy, near to state of art on both benchmarks.