# Identifying and Categorizing Offensive Language in Social Media

arXiv: https://arxiv.org/abs/2104.04871 · HTML: https://mlsanity.com/paper/2104.04871/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 3 of its 9 references that are in the map.
- First posted on arXiv 10 Apr 2021.

## Builds on (most cited first, 3 of 3)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Deep contextualized word representations](https://mlsanity.com/paper/1802.05365/) (2018)
- [SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval)](https://mlsanity.com/paper/1903.08983/) (2019)

## Built on by (most cited first, 1 of 1)

- [Obfuscation Rules for Detecting and Detoxifying Korean Toxicity](https://mlsanity.com/paper/2510.10961/) (2025)

## Abstract (arXiv)

Offensive language is pervasive in social media. Individuals frequently take advantage of the perceived anonymity of computer-mediated communication, using this to engage in behavior that many of them would not consider in real life. The automatic identification of offensive content online is an important task that has gained more attention in recent years. This task can be modeled as a supervised classification problem in which systems are trained using a dataset containing posts that are annotated with respect to the presence of some form(s) of abusive or offensive content. The objective of this study is to provide a description of a classification system built for SemEval-2019 Task 6: OffensEval. This system classifies a tweet as either offensive or not offensive (Sub-task A) and further classifies offensive tweets into categories (Sub-tasks B & C). We trained machine learning and deep learning models along with data preprocessing and sampling techniques to come up with the best results. Models discussed include Naive Bayes, SVM, Logistic Regression, Random Forest and LSTM.