An Empirical Comparison of Probability Models for Dependency Grammar

TL;DR

This paper empirically compares different probability models for dependency grammar parsing, demonstrating improved accuracy with larger training data and exploring variants that outperform previous models in attachment accuracy.

Contribution

It provides a detailed experimental comparison of dependency grammar models with larger training data, showing improved parsing accuracy and analyzing model differences.

Findings

Nearly half the sentences parsed with no misattachments

Two-thirds of sentences with at most one misattachment

Best model achieves 93% attachment accuracy with known tags

Abstract

This technical report is an appendix to Eisner (1996): it gives superior experimental results that were reported only in the talk version of that paper. Eisner (1996) trained three probability models on a small set of about 4,000 conjunction-free, dependency-grammar parses derived from the Wall Street Journal section of the Penn Treebank, and then evaluated the models on a held-out test set, using a novel O(n^3) parsing algorithm. The present paper describes some details of the experiments and repeats them with a larger training set of 25,000 sentences. As reported at the talk, the more extensive training yields greatly improved performance. Nearly half the sentences are parsed with no misattachments; two-thirds are parsed with at most one misattachment. Of the models described in the original written paper, the best score is still obtained with the generative (top-down) "model C." However, slightly better models are also explored, in particular, two variants on the comprehension (bottom-up) "model B." The better of these has an attachment accuracy of 90%, and (unlike model C) tags words more accurately than the comparable trigram tagger. Differences are statistically significant. If tags are roughly known in advance, search error is all but eliminated and the new model attains an attachment accuracy of 93%. We find that the parser of Collins (1996), when combined with a highly-trained tagger, also achieves 93% when trained and tested on the same sentences. Similarities and differences are discussed.