Re-estimation of lexical parameters for treebank PCFGs

Abstract

We present procedures which pool lexical information estimated from unlabeled data via the Inside-Outside algorithm, with lexical information from a treebank PCFG. The procedures produce substantial improvements (up to 31.6% error reduction) on the task of determining subcategorization frames of novel verbs, relative to a smoothed Penn Treebank-trained PCFG. Even with relatively small quantities of unlabeled training data, the re-estimated models show promising improvements in labeled bracketing f-scores on Wall Street Journal parsing, and substantial benefit in acquiring the subcategorization preferences of low-frequency verbs. © 2008 Licensed under the Creative Commons.