The goal of this project is to determine if we can predict the claim status (Yes or No) from the various travel insurance-related attributes.
The dataset has 63326 records. There are 11 fields/attributes where "Claim" is the target variable. Target: • Claim Status (Claim.Status) • Name of agency (Agency) • Type of travel insurance agencies (Agency.Type) • Distribution channel of travel insurance agencies (Distribution.Channel) • Name of the travel insurance products (Product.Name) • Duration of travel (Duration) • Destination of travel (Destination) • Amount of sales of travel insurance policies (Net.Sales) • Commission received for travel insurance agency (Commission) • Gender of insured (Gender) • Age of insured (Age)
More focus of the project was on EDA rather than Model Building. Since the Dataset was highly imbalanced dataset we have tried oversampling,undersampling,SMOTE to deal with it .