I’m a Computer Science master’s student, currently conducting my master thesis research on bias mitigation techniques for natural language processing, as well as Fair, Accountable and Transparent AI.
My field of work is Natural Language Processing and Responsible AI, particularly bias detection and mitigation on word embeddings.
Most of the literature and techniques developed for mitigating or detecting harmful biases in word embeddings come from the global north and are designed for English. Most of these techniques cannot be applied directly to Spanish, Portuguese or other languages. Moreover, cultural differences should be taken into consideration. This work is related to Feminist AI Research because its principal objective is to provide tools for developers to assess potential gender, race, and other biases in NLP models.
