dictNN: A Dictionary-Enhanced CNN Approach for Classifying Hate Speech on Twitter

TL;DR

This paper introduces dictNN, a CNN-based model enhanced with a crowd-sourced hate word dictionary, significantly improving hate speech detection accuracy on Twitter by leveraging continuous updates and fusion with standard embeddings.

Contribution

The paper presents a novel dictionary-enhanced vectorization method combined with CNNs for more reliable hate speech classification on social media.

Findings

F1 macro score increased by 7 percentage points with dictionary enhancement.

Model trained on a merged dataset of over 110,000 tweets.

Dictionary-based approach improves detection performance.

Abstract

Hate speech on social media is a growing concern, and automated methods have so far been sub-par at reliably detecting it. A major challenge lies in the potentially evasive nature of hate speech due to the ambiguity and fast evolution of natural language. To tackle this, we introduce a vectorisation based on a crowd-sourced and continuously updated dictionary of hate words and propose fusing this approach with standard word embedding in order to improve the classification performance of a CNN model. To train and test our model we use a merge of two established datasets (110,748 tweets in total). By adding the dictionary-enhanced input, we are able to increase the CNN model's predictive power and increase the F1 macro score by seven percentage points.