This report aims to provide a robust predictive model to predict the health outcome of horses in the Kaggle Playground Series - Season 3, Episode 22 competition. We experimented with well-known machine learning algorithms, including Random Forest, XGBoost, LightGBM, HistGradientBoosting, and CatBoost along with ensembling techniques such as stacking, to leverage the strengths of individual models. The performances of these model architectures were accessed in Kaggle to retrieve the official public leaderboard rankings and scores. Finally, to conclude the report, further insights on the evaluation of the effectiveness of different machine learning and ensembling architectures.
The project video can be viewed: https://youtu.be/NH_AlmmkIfo
We were ranked top 6% in the kaggle competition and our presentation was awarded the best submission in the corhot.