Excerpt from Towards Explainable, Fine-Grain Online Sexism Classification

15 Apr 2023

Methods to identify and classify sexist language online, using machine learning models like Gaussian Naive Bayes, bi-LSTM, and Transformer-based models. It achieved notable results with the RoBERTa-large model, placing in the top 10% for a specific task (SemEval-2023). The study aims to improve explainability in AI models for better understanding complex social issues like online sexism.


1. Introduction

3. Methods

4. Experiments

5. Results

6. Conclusion

7. Future Work

More details can be found in the report

Authors: Axel Bogos and Jie Bao, Université de Montréal