Hazard Count Prediction for Liberty Mutual’s Properties

Kaggle Ranking: Top 12% (248/2,362 competitors)

End Date: August 28, 2015

Description: Predict hazard counts for a house given anonymized variables such as roof, foundation, windows, and sliding. The goal is to help Liberty Mutual determine whether they should insure the property. Therefore, aligning Liberty Mutual’s portfolio of home insurance policies with their business goals.

Best Method: Weighted Ensemble (RF + XGBoost)

Reflections: This competition was particularly challenging because the anonymized features left little room for feature engineering. It was a fun exercise in ensembling and performing proper cross-validation to prevent overfitting on the public leaderboard, as tempting as that may be.