Location
Innovation Lab Atrium
Start Date
4-2-2025 2:00 PM
End Date
4-2-2025 3:30 PM
Presentation Date
2 April 2025, 2:00pm - 3:30pm
Biography
Djidjev biography:
Alex Djidjev is a second-year student majoring in Computer Science and Mathematics, graduating in December 2026. He is a Kummer Vanguard Scholar and an Honors Academy Scholar with a keen interest in the fields of Machine Learning and Data Science. He actively engages in research related to these areas, participating in the 2024-2025 OURE Program. Furthermore, he is part of the ACM Data organization, where he enjoys collaborating on projects with others and learning more about the field of data science. In his free time, he balances his academic pursuits with playing sports, running, and playing the piano.
Hellwege biography:
I am a senior Applied Mathematics major emphasizing in Statistics and Data Science. I grew up in St Louis, but I lived in Canada for about 6 years prior to coming here for school. I have always had a fascination with sports, baseball especially, and I will be starting a masters in Sports Analytics and Exercise Science at Marquette University in the fall. I am an officer in four different registered student organizations, including the Foundation for Undergraduate Mathematicians which I was a founding member of. I am also both a Kummer Vanguard scholar and a member of the honors academy.
Meeting Name
2025 - Miners Solving for Tomorrow Research Conference
Department(s)
Engineering Management and Systems Engineering
Document Type
Poster
Document Version
Final Version
File Type
event
Language(s)
English
Rights
© 2025 The Authors, All rights reserved
Predicting Baseball Game Wins with Machine Learning
Innovation Lab Atrium
Comments
Advisor: Gabriel Nicolosi
Abstract:
Baseball is a sport characterized by a multitude of statistics, each designed to capture different facets of the game. Due to the increasing application of machine learning and deep learning models in predictive tasks and the vast amount of data present in the sport of baseball, this study aims to investigate, compare, and enhance existing models for predicting the outcome of baseball games. We conducted a survey of current models, analyzed advanced baseball statistics, and collected relevant data. To capture the highly complex relationships, present in a baseball game, we utilized a feedforward neural network model, trained for a given team, with data from a variable number of preceding games to predict the next-game winner. Results indicate that further work will be necessary to formalize and test different neural network architectures to better understand the temporal aspects of an MLB team’s game performance.