Ghmerti at SemEval-2019 Task 6: A Deep Word- and Character-based Approach to Offensive Language Identification

TL;DR

This paper describes Ghmerti's deep learning models combining word and character features for offensive language detection in social media, achieving competitive results in SemEval-2019.

Contribution

It introduces a novel combination of character-level CNN and word-level RNN models with preprocessing for offensive language identification.

Findings

Achieved 77.93% macro F1-score on subtask A

Demonstrated effectiveness of combined character and word models

Provided insights into social media offensive language detection

Abstract

This paper presents the models submitted by Ghmerti team for subtasks A and B of the OffensEval shared task at SemEval 2019. OffensEval addresses the problem of identifying and categorizing offensive language in social media in three subtasks; whether or not a content is offensive (subtask A), whether it is targeted (subtask B) towards an individual, a group, or other entities (subtask C). The proposed approach includes character-level Convolutional Neural Network, word-level Recurrent Neural Network, and some preprocessing. The performance achieved by the proposed model for subtask A is 77.93% macro-averaged F1-score.