From Semantic Roles to Opinion Roles: SRL Data Extraction for Multi-Task and Transfer Learning in Low-Resource ORL

TL;DR

This paper develops a high-quality SRL dataset from OntoNotes for Opinion Role Labeling, enabling multi-task and transfer learning in low-resource opinion mining.

Contribution

It introduces a reproducible extraction pipeline that converts SRL annotations into an ORL-compatible dataset, facilitating research in low-resource opinion analysis.

Findings

Created a dataset with 97,169 predicate-argument instances

Aligned SRL roles with ORL schema for improved annotation

Provided detailed algorithms and analysis for dataset construction

Abstract

This report presents a detailed methodology for constructing a high-quality Semantic Role Labeling (SRL) dataset from the Wall Street Journal (WSJ) portion of the OntoNotes 5.0 corpus and adapting it for Opinion Role Labeling (ORL) tasks. Leveraging the PropBank annotation framework, we implement a reproducible extraction pipeline that aligns predicate-argument structures with surface text, converts syntactic tree pointers to coherent spans, and applies rigorous cleaning to ensure semantic fidelity. The resulting dataset comprises 97,169 predicate-argument instances with clearly defined Agent (ARG0), Predicate (REL), and Patient (ARG1) roles, mapped to ORL's Holder, Expression, and Target schema. We provide a detailed account of our extraction algorithms, discontinuous argument handling, annotation corrections, and statistical analysis of the resulting dataset. This work offers a reusable resource for researchers aiming to leverage SRL for enhancing ORL, especially in low-resource opinion mining scenarios.