Human-in-the-Loop for Data Collection: a Multi-Target Counter Narrative Dataset to Fight Online Hate Speech

TL;DR

This paper introduces a human-in-the-loop methodology for creating high-quality, diverse hate speech and counter narrative datasets, resulting in the first expert-based multi-target HS/CN dataset to combat online hate speech.

Contribution

It presents a novel iterative data collection approach using generative models refined by experts, improving dataset quality and diversity for counter narrative generation.

Findings

Method is scalable and cost-effective.

Produces diverse and novel counter narratives.

Results in the only expert-based multi-target HS/CN dataset.

Abstract

Undermining the impact of hateful content with informed and non-aggressive responses, called counter narratives, has emerged as a possible solution for having healthier online communities. Thus, some NLP studies have started addressing the task of counter narrative generation. Although such studies have made an effort to build hate speech / counter narrative (HS/CN) datasets for neural generation, they fall short in reaching either high-quality and/or high-quantity. In this paper, we propose a novel human-in-the-loop data collection methodology in which a generative language model is refined iteratively by using its own data from the previous loops to generate new training samples that experts review and/or post-edit. Our experiments comprised several loops including dynamic variations. Results show that the methodology is scalable and facilitates diverse, novel, and cost-effective data collection. To our knowledge, the resulting dataset is the only expert-based multi-target HS/CN dataset available to the community.