KaterynaD / Water-Peril-Claims-Research-with-XGBoost-and-GLM-models

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Water-Peril-Claims-Research-with-XGBoost-and-GLM-models

The original main purpose of the project is to build several machine learning models using different methods, compare results and estimate usability for analyzing risks. Property modeling data were recently built and look like an easy (well, at least, smaller) dataset then Auto modeling data. Adding in the system LocationInc water risks scores narrowed the interest to water peril only. Discovering the power of Shap Values and Lorenz curve added additional purposes to visualizing models results in Tableau dashboards. Precise feature analysis, model tuning and evaluation are not the purpose of the project and not included in the results.

About


Languages

Language:HTML 62.7%Language:Jupyter Notebook 36.3%Language:Python 1.0%Language:R 0.0%