Semantic Role Labeling of NomBank Partitives

TL;DR

This paper explores semantic role labeling for English partitive nouns in NomBank, employing traditional and transformer-based models, achieving high accuracy with ensemble methods in both classroom and experimental settings.

Contribution

It introduces systems for semantic role labeling of partitive nouns using advanced machine learning techniques, including transformer models and ensembling, with state-of-the-art performance.

Findings

Highest F1 score of 91.74% with gold parses

Achieved 91.12% F1 with neural parser

Demonstrated effectiveness in classroom and experimental settings

Abstract

This article is about Semantic Role Labeling for English partitive nouns (5%/REL of the price/ARG1; The price/ARG1 rose 5 percent/REL) in the NomBank annotated corpus. Several systems are described using traditional and transformer-based machine learning, as well as ensembling. Our highest scoring system achieves an F1 of 91.74% using "gold" parses from the Penn Treebank and 91.12% when using the Berkeley Neural parser. This research includes both classroom and experimental settings for system development.