← Back to ProjectsGitHub
UFC Fight Prediction Model
Built an end-to-end ML pipeline to predict fight outcomes and evaluate which features and models generalize best.
PythonPandasScikit-learnFeature EngineeringEvaluation
Role
Data Scientist / ML Engineer (Project)
Timeline
2025
Stack
Python • Pandas • Scikit-learn • Feature Engineering • Cross-validation
Goal
Predict fight winner
Approach
Feature selection + model comparison
Focus
Generalization & evaluation
What I did
- Built a repeatable training pipeline (data cleaning → feature engineering → train/validate → evaluation).
- Compared multiple models and tuned hyperparameters to improve predictive performance.
- Tested feature selection strategies to identify the most informative signals and reduce noise/overfitting.
- Tracked results with consistent metrics and validation to avoid misleading conclusions.