arnabd1987 / Classifier-Driver-Analysis---Market-Research-

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Classifier-Driver-Analysis---Market-Research-

#User Guide

DriverAnalysis_C is the main program

All the *.py files in the reporsitory has to be in the same folder for this proram to run

Keep the data file in the same folder

Sample Size should be more than number of independent varivale * 20 at bare minimum. If there are 10 independent variable then sample size should be 10*20=200

Data file has to be in CSV format

Ensure dependent variable is the last variable of the data file

There should not be any index column. Data file should have only dependent and independent variable

Can have n mumber of independent variable

This repository is based on Classification algorithm. Hence it will only work when independent variable is binary

Input data has to be processed

Ensure all nominal data (Age, Gender, marital status etc.) has been converted to binary varibale

All ordinal data has to be recoded in the based of their importance(i.e. NCCS A=5, NCCS B=4...)

Remove any variable with high order of correlation

#Consider

About