14 - Decision Trees.txt

Problem 01
About the data: 
Let’s consider a Company dataset with around 10 variables and 400 records. 
The attributes are as follows: 
 Sales -- Unit sales (in thousands) at each location
 Competitor Price -- Price charged by competitor at each location
 Income -- Community income level (in thousands of dollars)
 Advertising -- Local advertising budget for company at each location (in thousands of dollars)
 Population -- Population size in region (in thousands)
 Price -- Price company charges for car seats at each site
 Shelf Location at stores -- A factor with levels Bad, Good and Medium indicating the quality of the shelving location for the car seats at each site
 Age -- Average age of the local population
 Education -- Education level at each location
 Urban -- A factor with levels No and Yes to indicate whether the store is in an urban or rural location
 US -- A factor with levels No and Yes to indicate whether the store is in the US or not
The company dataset looks like this: 
 
Problem Statement:
A cloth manufacturing company is interested to know about the segment or attributes causes high sale. 
Approach - A decision tree can be built with target variable Sale (we will first convert it in categorical variable) & all other variable will be independent in the analysis.  


Problem 02
Use decision trees to prepare a model on fraud data 
treating those who have taxable_income <= 30000 as "Risky" and others are "Good"

Data Description :

Undergrad : person is under graduated or not
Marital.Status : marital status of a person
Taxable.Income : Taxable income is the amount of how much tax an individual owes to the government 
Work Experience : Work experience of an individual person
Urban : Whether that person belongs to urban area or not