Calibrating Features for Semantic Role Labeling

Nianwen Xue, Martha Palmer

2004 · 322 citations · 15 references

Concepts

TL;DR

The paper critiques the features used in semantic role tagging, noting that input information from syntactic parse trees is not fully exploited. The authors propose an additional set of features for semantic role labeling. They evaluate these features through experiments that demonstrate significant improvements in the tasks performed. The experiments reveal that different features are needed for different subtasks, that a Maximum Entropy classifier with fewer features can match the best SVM results, and that selecting features that capture the right information is essential for advancing semantic analysis.

Abstract

This paper takes a critical look at the features used in the semantic role tagging literature and show that the information in the input, generally a syntactic parse tree, has yet to be fully exploited. We propose an additional set of features and our experiments show that these features lead to fairly significant improvements in the tasks we performed. We further show that different features are needed for different subtasks. Finally, we show that by using a Maximum Entropy classifier and fewer features, we achieved results comparable with the best previously reported results obtained with SVM models. We believe this is a clear indication that developing features that capture the right kind of information is crucial to advancing the stateof-the-art in semantic analysis.

References

15