I am a Ph.D. candidate in Political Science at McGill University. I specialize in developing ways to use language models to understand political language, especially in international diplomacy. As a secondary interest, I also study global governance of artificial intelligence. Additionally, I have done some advisory work on AI and pedagogy at McGill, and have taught some methodological workshops on how to incorporate language models as a research tool.
- Domesticating Wealth Inequality. Vincent Pouliot and Scott Robert Patterson. Global Studies Quarterly, 2024.
- Placing Machine Learning into the Hermeneutic Circle. Scott Robert Patterson and Vincent Pouliot. Journal of International Relations and Development, 2024.
- State Detection NER Demo - This is a custom named entity detection (NER) model that can extract mentions of countries from unstructured text. Follow the link to test out the most recent version.
- Occupation Classifier - A fine-tuned GPT model for classifying occupations using BLS codes. Follow the link to try the most recent version.
- Audio transcription - This illustrates how to transcribe audio files for free using the OpenAI whisper model. If working with many files or with large files, this colab version works better and allows for free-tier GPU acceleration.
Contact: scott[dot]patterson[at]mail[dot]mcgill[dot]ca