Descriptive and empirical approaches to capturing underlying dependencies among parsing errors

Abstract

In this paper, we provide descriptive and empirical approaches to effectively extracting underlying dependencies among parsing errors. In the descriptive approach, we define some combinations of error patterns and extract them from given errors. In the empirical approach, on the other hand, we re-parse a sentence with a target error corrected and observe errors corrected together. Experiments on an HPSG parser show that each of these approaches can clarify the dependencies among individual errors from each point of view. Moreover, the comparison between the results of the two approaches shows that combining these approaches can achieve a more detailed error analysis. © 2009 ACL and AFNLP.