# Enhanced Offensive Language Detection Through Data Augmentation

arXiv: https://arxiv.org/abs/2012.02954 · HTML: https://mlsanity.com/paper/2012.02954/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Dec 2020.

## Built on by (most cited first, 4 of 4)

- [A Survey of Data Augmentation Approaches for NLP](https://mlsanity.com/paper/2105.03075/) (2021)
- [Offense Detection in Dravidian Languages using Code-Mixing Index based Focal Loss](https://mlsanity.com/paper/2111.06916/) (2021)
- [BLIAM: Literature-based Data Synthesis for Synergistic Drug Combination Prediction](https://mlsanity.com/paper/2302.06860/) (2023)
- [Ensemble of pre-trained language models and data augmentation for hate speech detection from Arabic tweets](https://mlsanity.com/paper/2407.02448/) (2024)

## Abstract (arXiv)

Detecting offensive language on social media is an important task. The ICWSM-2020 Data Challenge Task 2 is aimed at identifying offensive content using a crowd-sourced dataset containing 100k labelled tweets. The dataset, however, suffers from class imbalance, where certain labels are extremely rare compared with other classes (e.g, the hateful class is only 5% of the data). In this work, we present Dager (Data Augmenter), a generation-based data augmentation method, that improves the performance of classification on imbalanced and low-resource data such as the offensive language dataset. Dager extracts the lexical features of a given class, and uses these features to guide the generation of a conditional generator built on GPT-2. The generated text can then be added to the training set as augmentation data. We show that applying Dager can increase the F1 score of the data challenge by 11% when we use 1% of the whole dataset for training (using BERT for classification); moreover, the generated data also preserves the original labels very well. We test Dager on four different classifiers (BERT, CNN, Bi-LSTM with attention, and Transformer), observing universal improvement on the detection, indicating our method is effective and classifier-agnostic.