Soccer Analysis Dashboard — McGill Hackathon
End-to-end pipeline to visualize soccer player metrics and performance insights using R Shiny and web scraping. Presented the final tool to a panel of judges — earned 1st place at the competition.
Applied Data Science • Analytics • Sports Analytics
I'm a data scientist who loves turning messy, real-world data into things people can actually use — whether that's a competition-winning sports dashboard, a smarter operations tool, or a bot-detection pipeline. I care about the full stack: from cleaning raw data to shipping something clear and useful.
I'm a Master's student in Applied Data Science at the University of Chicago (Expected Aug 2026), building on a B.Sc. in Statistics & Computer Science from McGill (2021–2025). My background combines rigorous statistical theory with hands-on software engineering, giving me a strong foundation across machine learning, statistical inference, and data engineering.
I'm particularly drawn to projects where data directly informs decisions — optimizing systems, understanding human behavior, or improving performance in applied settings like sports analytics. Through internships, hackathons, and independent projects, I've learned the importance of writing maintainable code, validating assumptions, and presenting results in ways that both technical and non-technical audiences can act on.
Selected work (more on GitHub).
End-to-end pipeline to visualize soccer player metrics and performance insights using R Shiny and web scraping. Presented the final tool to a panel of judges — earned 1st place at the competition.
Streamlit dashboard that pulls Strava activity data and analyzes training load, fatigue/readiness, and race predictions using explainable rules + time-series analytics.
Web application to enhance safety awareness and route planning in Montreal. Uses a map interface with hex overlays representing safety zones to help users choose routes based on safety considerations.
Applied NLP and cosine similarity-based clustering alongside feature engineering (profanity, emoji detection, grammar signals) to identify bots injected into a large tweet dataset. Achieved ~90% accuracy.
Streamlit educational app to explore asteroid impact scenarios with "confidence level" guardrails, plus optional NASA feeds integration for current events/context.
Analysis of Taylor Swift's media presence using web scraping and NLP to identify trends and patterns. For details, see the written report in the repo.
Recent roles and impact.
Sep 2024 – Aug 2025
May 2024 – Aug 2024
Jul 2022
Selected advanced coursework and applied competitions.
University of Chicago
McGill University
Best way to reach me:
© Arthur Acker • Built with HTML & CSS