Michaela Drouillard
hi!
I’m into data science, legal applications, governance and cultural analytics. I’m also into modelling style in art, writing, and music using computational methods.
I’m a Faculty Affiliate Researcher at the Vector Institute, and my research is supported by the Data Sciences Institute at the University of Toronto. My supervisors are Tegan Maharaj and Rohan Alexander. I’m a co-organizer of the Toronto Data Workshop with Rohan Alexander and Kelly Lyons.
M.I. Data Science, University of Toronto (August 2024)
- I modelled stylistic variation in pop music production, which started out as a casual investigation into Jack Antonoff's style as described by the Spotify API's audio features. Then I reproduced the API's “danceability” feature to see what’s going on under the hood there, a process informed by interviews with Spotify principal engineers and my undergraduate training in cultural analytics.
B.A. Classics and Russian Literature, McGill University (2016-2021)
- I translated Euripides’ Medea from Ancient Greek to English and then directed an original adaptation of it with my friend Marina Martin. We were awarded the Paul F McCullagh Award by the Faculty of Arts at McGill.
📩michaela.drouillard@mail.utoronto.ca
Research links + Projects:
Drouillard, M., Spencer, R., Allen, N., Maharaj, T. (2024) "Quantifying Likeness: A Simple Machine Learning Approach to Identifying Copyrighted Elements of Style in (AI-Generated) Artwork". Proceedings of the 2nd GenLaw workshop at the 2024 International Conference on Machine Learning (ICML). https://www.genlaw.org/2024-icml-papers#quantifying-likeness-a-computer-vision-approach-to-identifying-style-and-copyright-infringement-in-ai-generated-artwork *
Alexander, R., Katz, L., Moore, C., Drouillard, M., Wing-Cheung Wong, M., Schwartz, Z. (2024). “Evaluating the Decency and Consistency of Data Validation Tests Generated by LLMs: An application to Canadian political donations data”. https://arxiv.org/html/2310.01402v2
Erlin, M., Piper, A., Knox, D., Pentecost, S., Drouillard, M., Townson, C. (2021). “Cultural Capitals: Modeling ‘Minor’ European Literature.” Journal of Cultural Analytics. 44-77. DOI: 10.7910/DVN/SDMGRC
I also built the web scrapers responsible for the Investigative Journalism Foundation’s lobbying registrations database.
* I will be giving a talk about this project at the 2024 Montreal AI Symposium on AI and governance!
Tech stack:
Programming: Python, R
Libraries & Frameworks: pandas, numpy, scikitlearn, pyspark, plotly, StanfordNLP, BeautifulSoup, lxml, MESA
Databases: SQL (Postgres)
Platforms: Azure, Databricks
Other:
I like running, hanging out at baseball games, napping on the dock, reading (novels, plays, books about businesses, biographies), and doing the Atlantic crossword.