Welcome to the repository for the Gender Bias Analysis project presented at the KSE 2023 conference. This repository contains the code and resources used for our research on uncovering and mitigating gender bias in commonsense knowledge graph embeddings.
Knowledge graphs play a fundamental role in various natural language understanding tasks, providing a structured representation of information. However, biases present in knowledge graphs can have detrimental effects on downstream applications. Gender bias, in particular, is a significant concern, as it can reinforce and perpetuate stereotypes in AI systems. Our project aims to address this issue by analyzing and mitigating gender bias in commonsense knowledge graph embeddings.
Please cite the following paper if it helps your research:
- Khine Myat Thwe, Teeradaj Racharak, Minh Le Nguyen, Gender Bias Analysis in Commonsense Knowledge Graph Embeddings, In Proceedings of the 15th IEEE International Conference on Knowledge and Systems Engineering (KSE), Ha Noi, Vietnam, October 18-20, 2023