KMeans Clustering-Based Feature Engineering
Experimented with KMeans clustering as a feature engineering technique to enhance model performance in a Kaggle tabular competition. Observed improvements in both CV and LB scores by combining clustering features with domain-driven ones like BMI, BMR, and Intensity.
KMeans
scikit-learn
XGBoost
Pandas
Python
#feature-engineering
#clustering
#kaggle
🔗 See Project