Question
In this project, you are required to build a model to identify inputs or predictors that differentiate risky from non-risky clients (based on patterns about
In this project, you are required to build a model to identify inputs or predictors that differentiate risky from non-risky clients (based on patterns about previous clients) in an insurance company. You are asked to use those inputs to predict new customers who might be charged at a high, medium, or low premium. A sample data file can be found on Blackboard. The data file consists of 490 cases and 14 variables about past and current clients of the insurance company. The data set contains client-related information such as vehicle type, ownership type, years of owning a vehicle, employment, demographic data, and the outcome or dependent variable for each client, classifying each case as low, medium, or high based on the institution's past data. In the data file, 490 data points can be found under the Excel Sheet labeled Insurance Sample File. Using the sample data points in the attached file,
- Build a decision tree model consisting of three independent variables that will best provide you with insights to learn about the characteristics of the problem. Test its performance on the 15 data points listed under New Data, found at the bottom of the Sheet.
- Create two additional decision tree models, using three independent variables different from the ones used to create the first decision tree model.
- Compare the three models and report following the CRISP-DM approach. Your report should include a discussion on the resulting performance of the test set (i.e., New Data) and recommendations to an Insurance Manager who needs to use your model to determine customers who should be charged low medium, or high premiums. Be sure to include any assumptions.
New Data | |||||||||||||
Pickup | 2004 | Lease | Personal Injury | Motorcycle | 5 | 33 | M | Divorced | 69 | 1 | 1 | Unemployed | |
Crossover | 2010 | Owner | Comprehensive Coverage | Business | 7 | 71 | F | Divorced | 56 | 3 | 0 | Management | |
SUV | 2019 | Lease | Liability Coverage | Boat | 20 | 139 | F | Divorced | 21 | 1 | 1 | Management | |
Sedan | 2016 | Lease | Medical Coverage | Landlord | 2 | 44 | F | Married | 55 | 2 | 5 | Management | |
Pickup | 2012 | Owner | Comprehensive Coverage | Flood | 13 | 43 | M | Married | 43 | 5 | 3 | Unemployed | |
Pickup | 2020 | Lease | Collision Coverage | Motorcycle | 2 | 107 | M | Single | 72 | 3 | 3 | Skilled | |
Sedan | 2016 | Owner | Uninsured | Boat | 6 | 79 | F | Married | 20 | 5 | 3 | Skilled | |
Wagon | 2006 | Owner | Collision Coverage | Condo | 20 | 49 | F | Single | 48 | 3 | 0 | Skilled | |
Wagon | 2018 | Owner | Uninsured | Boat | 12 | 120 | F | Married | 34 | 0 | 3 | Management | |
Wagon | 2016 | Lease | Liability Coverage | Business | 1 | 85 | M | Married | 52 | 4 | 4 | Unemployed | |
Pickup | 2017 | Lease | Medical Coverage | Motorcycle | 12 | 117 | M | Divorced | 40 | 2 | 0 | Skilled | |
Minivan | 2015 | Owner | Collision Coverage | Condo | 3 | 134 | F | Divorced | 16 | 4 | 0 | Unemployed | |
Sedan | 2019 | Lease | Comprehensive Coverage | Flood | 9 | 131 | M | Married | 19 | 1 | 1 | Management | |
SUV | 2015 | Lease | Medical Coverage | Landlord | 19 | 60 | F | Divorced | 54 | 2 | 5 | Unemployed | |
Pickup | 2011 | Owner | Medical Coverage | Flood | 19 | 115 | M | Married | 17 | 2 | 1 | Skilled |
Step by Step Solution
There are 3 Steps involved in it
Step: 1
Get Instant Access to Expert-Tailored Solutions
See step-by-step solutions with expert insights and AI powered tools for academic success
Step: 2
Step: 3
Ace Your Homework with AI
Get the answers you need in no time with our AI-driven, step-by-step assistance
Get Started