Improve Parsing Performance by Self-Learning

There are many methods to improve performance of statistical parsers. Resolving structural ambiguities is a major task of these methods. In the proposed approach, the parser produces a set of n-best trees based on a feature-extended PCFG grammar and then selects the best tree structure based on association strengths of dependency word-pairs. However, there is no sufficiently large Treebank producing reliable statistical distributions of all word-pairs. This paper aims to provide a self-learning method to resolve the problems. The word association strengths were automatically extracted and learned by parsing a giga-word corpus. Although the automatically learned word associations were not perfect, the constructed structure evaluation model improved the bracketed f-score from 83.09% to 86.59%. We believe that the above iterative learning processes can improve parsing performances automatically by learning word-dependence information continuously from web.