HBert + BiasCorp -- Fighting Racism on the Web

TL;DR

This paper introduces hBERT, a modified BERT model with Hopfield Layers, and BiasCorp, a dataset of racially biased comments, to combat online racism through NLP tools and resources.

Contribution

It presents a novel BERT variant with Hopfield Layers and a large annotated dataset for detecting racial bias online, along with practical tools for deployment.

Findings

hBERT generalizes well across different data distributions

BiasCorp dataset contains 139,090 comments with 45,000 manually annotated

Tools enable detection and reporting of racial bias in web content

Abstract

Subtle and overt racism is still present both in physical and online communities today and has impacted many lives in different segments of the society. In this short piece of work, we present how we're tackling this societal issue with Natural Language Processing. We are releasing BiasCorp, a dataset containing 139,090 comments and news segment from three specific sources - Fox News, BreitbartNews and YouTube. The first batch (45,000 manually annotated) is ready for publication. We are currently in the final phase of manually labeling the remaining dataset using Amazon Mechanical Turk. BERT has been used widely in several downstream tasks. In this work, we present hBERT, where we modify certain layers of the pretrained BERT model with the new Hopfield Layer. hBert generalizes well across different distributions with the added advantage of a reduced model complexity. We are also releasing a JavaScript library and a Chrome Extension Application, to help developers make use of our trained model in web applications (say chat application) and for users to identify and report racially biased contents on the web respectively.