Imbalanced learning for insurance using modified loss functions in tree-based models

Keyphrases

• Tree Model 100%
• Insurance 100%
• Zero Responses 100%
• Imbalanced Learning 100%
• Loss Modeling 100%
• Modified Loss Function 100%
• Performance Prediction 66%
• Insurance Losses 66%
• Splitting Function 66%
• Tree Structure 33%
• Structure Prediction 33%
• Synthetic Data 33%
• Weighted Sums 33%
• Tree Algorithm 33%
• Traditional Models 33%
• Heavy Tails 33%
• Loss Distribution 33%
• Improved Prediction 33%
• Splitting Method 33%
• Sum Square Error 33%
• Point Mass 33%
• Classification and Regression Tree 33%
• Insurance Claims 33%
• Canberra 33%

Mathematics

• Loss Function 100%
• Loss Modeling 100%
• Squared Error 33%
• Weighted Sum 33%
• Tree Structure 33%
• Heavy Tail 33%
• Predictive Performance 33%
• Regression tree 33%
• Point Mass 33%
• Loss Distribution 33%
• Insurance Claim 33%
• Loss Insurance 33%