Combining Behavioral Research and NLP to Combat Gender Biases

Watch a 31-minute conference talk from Data Science Conference Europe 2022 exploring how AI can be leveraged to recognize and mitigate gender biases rather than perpetuate them. Learn how Ahmad Haj Mosa and Marcel Tkacik developed a full-stack application that integrates social science research with modern MLOps and responsible AI practices. Discover various techniques including short and long-term learning cycles from user interactions, few/one/multiple-shot learning methods, model confidence measures, and Kubernetes deployment strategies. Delivered in Belgrade, this presentation demonstrates the practical implementation of combining behavioral research with natural language processing to address and combat gender biases in artificial intelligence systems.