MLB Game-to-game Attendance Prediction for Minnesota Twins

Be part of MinneMUDAC 2023 Data Challenge, we developed different Machine Learning models (Decision Tree, Random Forest, XGBoost, kNN) to forecast Minnesota Twins’s home-game attendance. The process involved embedding capture variables relation in feature spaces to improve model performance. Our final result is a dynamic model using Cat-Boost Gradient Boosting with comparably low MAE score. We were able to identified trends and important factors, analyzed distributions, and provided impactful business insights to our clients (MINTwins).

STACKMachine Learning, Web-scraping, Data Modeling, Scikit-learn, Pandas, Matplotlib

Platformhttps://minneanalytics.org/minnemudac2023/

GITHUBhttps://github.com/vynpt/minnemudac2023