Collins-LA: Collins' Head-Driven Model with Latent Annotation
Seung‐Hoon Na, Meixun Jin, In-Su Kang, Jong-Hyeok Lee · 2008
Recent works on parsing have reported that the lexicalization does not have a serious role for parsing accuracy. Latent-annotation methods such as PCFG-LA are one of the most promising un-lexicalized approaches, and reached the-state-of-art performance. However, most works on latent annotation have investigated only PCFG formalism, without considering the Collins’ popular head-driven model, though it is a significantly important and interesting issue. To this end, this paper develops Collins-LA, the extension of the Collins’ head-driven model to support the latent annotation. We report its basic accuracy, comparing with PCFG-LA. The experimental results show that Collins-LA has potential to improve basic parsing accuracy, resulting in comparable performance with PCFG-LA even in the naive setting.