ProjectFlow.txt

Data:-
Divide Data in m-n spilt (m% for train and n% for test)
Select the Alogrithm to work on (should be drop down menu) for now only option is Decision Tree
Do AutoSegmentation on the Training set i.e get score for each features categories
Set Cutoff where we define the minimum population for the category to get selected if minimum population doesn't match then don't show that feature
Note:-keep this feature optional ask the user if its needs to setup a cutoff if yes then only use the cutoff feature
Select the best segmented Category manually
create a seperate dataframe with only data having value as in category for e.g if category is male create seperate df which only inclue male
and on another DF which include values except of selected category for e.g Male and perform the above STEPS from line 4 again by passing the excluded selected category df
perform this iteration till user manually say stop. if user say stop it means at that iteration user will not select any category.
so now lets say we have 4 iteration 1:-Male.2:-AGE>30.3:-Married.4:-rest of data(data when user told to stop)
Now train 4 decision tree on this 4 dataset
Now iterate the test dataset rows 1 by 1 in which if any category is from above 3 category then predict that category model or predict its value from the rest of data model