Stephen Scarano

Stephen Scarano

Data Scientist (M.S) with Interests in Artifical Intelligence

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

Uncommon Good Substack

less than 1 minute read

Published:

I am a grateful alumni of the Socially-Intelligent Media & Systems Lab (SIMs), where I primarily worked on analyzing social polling and leveraging participant characteristics to predict election outcomes. Blog posts about this work will eventually be posted on the lab’s substack, Uncommon Good.

portfolio

Energy Consumption Prediction for Climate Planning

For Data Science for the Common Good (DS4CG), produced traditional machine learning and Gaussian processes to forecast property emissions data for the Massachusetts Department of Capital Asset Management and Maintenance (DCAMM).

Analyzing Support for U.S. Presidential Candidates in Twitter Polls

We characterize social polls online and leverage machine learning models to describe the demographics, political leanings, and other characteristics of the users who author and interact with social polls. We study the relationship between social poll results, their attributes, and the characteristics of users interacting with them.

Robust Frame-to-Frame Camera Rotation Estimation in Crowded Scenes

We characterize social polls online and leverage machine learning models to describe the demographics, political leanings, and other characteristics of the users who author and interact with social polls. We study the relationship between social poll results, their attributes, and the characteristics of users interacting with them.

publications

Robust Frame-to-Frame Camera Rotation Estimation in Crowded Scenes

Published in International Conference on Computer Vision (ICCV), 2023

We introduce a novel generalization of the Hough transform on SO3 to efficiently find the camera rotation most compatible with optical flow.

Recommended citation: Fabien Delattre, David Dirnfeld, Phat Nguyen, Stephen Scarano, Michael J. Jones, Pedro Miraldo, and Erik Learned-Miller. 2023. Robust Frame-to-Frame Camera Rotation Estimation in Crowded Scenes https://arxiv.org/abs/2309.08588

Analyzing Support for U.S. Presidential Candidates in Twitter Polls

Published in Journal of Quantitative Description (JQD:DM) and the International AAAI Conference on Web and Social Media (ICWSM), 2024

We characterize social polls online and leverage machine learning models to describe the demographics, political leanings, and other characteristics of the users who author and interact with social polls. We study the relationship between social poll results, their attributes, and the characteristics of users interacting with them.

Recommended citation: Scarano, S. et al. 2024. Analyzing Support for U.S. Presidential Candidates in Twitter Polls. Journal of Quantitative Description: Digital Media . 4, (May 2024). DOI:https://doi.org/10.51685/jqd.2024.icwsm.4. https://journalqd.org/article/view/5897

Election Polls on Social Media: Prevalence, Biases, and Voter Fraud Beliefs

Published in International AAAI Conference on Web and Social Media (ICWSM), 2024

We leverage demographic inference and statistical analysis, finding that Twitter polls are disproportionately authored by older males, exhibit a large bias towards candidate Donald Trump relative to representative mainstream polls, and contain inconsistencies between public vote counts and those privately visible to poll authors.

Recommended citation: (DOI and citation coming soon) https://arxiv.org/abs/2405.11146

talks

teaching