Active Learning for Argument Mining: A Practical Approach

TL;DR

This paper demonstrates that active learning significantly reduces annotation effort while maintaining high performance in argument mining tasks, specifically for argument unit recognition and classification.

Contribution

It provides a large-scale comparison of active learning methods, showing their effectiveness in reducing resource creation costs in argument mining.

Findings

Active learning decreases annotation effort for argument mining.

Active learning achieves comparable performance to full data training.

Large-scale comparison of active learning methods in argument mining.

Abstract

Despite considerable recent progress, the creation of well-balanced and diverse resources remains a time-consuming and costly challenge in Argument Mining. Active Learning reduces the amount of data necessary for the training of machine learning models by querying the most informative samples for annotation and therefore is a promising method for resource creation. In a large scale comparison of several Active Learning methods, we show that Active Learning considerably decreases the effort necessary to get good deep learning performance on the task of Argument Unit Recognition and Classification (AURC).