Static and Dynamic Feature Selection in Morphosyntactic Analyzers

TL;DR

This paper explores feature selection methods for morphosyntactic analyzers, demonstrating that dynamic feature ordering improves accuracy and efficiency across multiple languages and system configurations.

Contribution

It introduces and compares static and dynamic feature selection techniques, showing dynamic ordering's superiority and applying these methods to enhance joint tagging and parsing systems.

Findings

Dynamic feature ordering outperforms static ordering.

Feature selection reduces model size and increases accuracy.

Joint systems benefit more from feature selection.

Abstract

We study the use of greedy feature selection methods for morphosyntactic tagging under a number of different conditions. We compare a static ordering of features to a dynamic ordering based on mutual information statistics, and we apply the techniques to standalone taggers as well as joint systems for tagging and parsing. Experiments on five languages show that feature selection can result in more compact models as well as higher accuracy under all conditions, but also that a dynamic ordering works better than a static ordering and that joint systems benefit more than standalone taggers. We also show that the same techniques can be used to select which morphosyntactic categories to predict in order to maximize syntactic accuracy in a joint system. Our final results represent a substantial improvement of the state of the art for several languages, while at the same time reducing both the number of features and the running time by up to 80% in some cases.