Choosing features for classifying multiword expressions

2026-05-12

AI Summary

This study addresses the challenge of classifying multiword expressions (MWEs), which exhibit high typological heterogeneity and lack a unified computational framework in linguistics. The key difficulty lies in identifying classification features that remain stable and effective across languages. To tackle this, the paper proposes an optimization approach grounded in feature reliability assessment, systematically quantifying the contribution of various prior features to MWE categorization. By integrating findings from multilingual research, the authors develop a classification framework that enhances both computational utility and cross-lingual adaptability. This framework overcomes limitations inherent in traditional taxonomies when applied to natural language processing tasks, thereby laying the groundwork for building high-quality, transferable linguistic resources.

Abstract

Multiword expressions (MWEs) are a heterogeneous set with a glaring need for classifications. Designing a satisfactory classification involves choosing features. In the case of MWEs, many features are a priori available. Not all features are equal in terms of how reliably MWEs can be assigned to classes. Accordingly, resulting classifications may be more or less fruitful for computational use. I outline an enhanced classification. In order to increase its suitability for many languages, I use previous works taking into account various languages.

Problem

• multiword expressions
• classification
• feature selection
• computational linguistics
• cross-lingual

Innovation

• multiword expressions
• feature selection
• cross-lingual classification
• computational linguistics
• enhanced classification