Detecting Online Hate Speech Using Context Aware Models

TL;DR

This paper introduces context-aware models for detecting online hate speech, leveraging additional contextual information to improve accuracy, and demonstrates significant performance gains over baseline methods.

Contribution

It provides an annotated hate speech dataset with context and proposes two novel models that incorporate context information for improved detection accuracy.

Findings

Both models outperform baseline by 3-4% in F1 score.

Combining models yields an additional 7% improvement.

Context information significantly enhances hate speech detection.

Abstract

In the wake of a polarizing election, the cyber world is laden with hate speech. Context accompanying a hate speech text is useful for identifying hate speech, which however has been largely overlooked in existing datasets and hate speech detection models. In this paper, we provide an annotated corpus of hate speech with context information well kept. Then we propose two types of hate speech detection models that incorporate context information, a logistic regression model with context features and a neural network model with learning components for context. Our evaluation shows that both models outperform a strong baseline by around 3% to 4% in F1 score and combining these two models further improve the performance by another 7% in F1 score.