In [124]:

    

import pandas as pd 
import numpy as np 
import matplotlib.pyplot as plt 

def cat_split(n):
    def inner(x):
        hierarchy = x.split('>')
        hierarchy = hierarchy + [""]*(n - len(hierarchy))
        return tuple(hierarchy)
    return inner


    

In [125]:

    

problems = pd.read_csv('problems.csv', sep=', ')
problems['Code'] = problems['Code'].apply(cat_split(3))
problems


    

    
Out[125]:






  

    

      

      
Problem
      
Pain
      
People
      
Code
    
  
  

    

      
0
      
Uses pact
      
1
      
johnf
      
(accountability, uses-product, )
    
    

      
1
      
Uses fitbit
      
2
      
johnf
      
(tracking, uses-product, )
    
    

      
2
      
Has read "The Power of Habit"
      
1
      
jake
      
(challenges, passive, reading)
    
    

      
3
      
Shares with his wife
      
1
      
jake
      
(accountability, wife, )
    
    

      
4
      
Asked GF to ring in the morning
      
2
      
johnf
      
(accountability, wife, )
    
  

In [115]:

    

people = pd.read_csv('people.csv', sep=', ')
people


    

    
Out[115]:






  

    

      

      
People
      
First Name
      
Last Name
      
Age
    
  
  

    

      
0
      
johnf
      
John
      
Fawcett
      
38
    
    

      
1
      
jake
      
Jake
      
Desyllas
      
43
    
  

In [126]:

    

demographics = pd.read_csv('demographics.csv', sep=', ')
demographics['Demographic'] = demographics['Demographic'].apply(cat_split(3))
demographics


    

    
Out[126]:






  

    

      

      
Demographic
      
People
    
  
  

    

      
0
      
(job, entrepreneur, )
      
johnf
    
    

      
1
      
(job, creative, blogger)
      
johnf
    
    

      
2
      
(job, creative, writer)
      
johnf
    
    

      
3
      
(habit, diet, )
      
johnf
    
    

      
4
      
(habit, exercise, )
      
johnf
    
    

      
5
      
(job, coach, )
      
jake
    
    

      
6
      
(job, entrepreneur, )
      
jake
    
    

      
7
      
(habit, writing, )
      
jake
    
    

      
8
      
(habit, wakeing-up, )
      
jake
    
  

In [148]:

    

prob_demo = pd.merge(problems, demographics, on='People')
prob_demo.stack()
prob_demo.unstack(1)


    

    
Out[148]:





Problem      0                           Uses pact
             1                           Uses pact
             2                           Uses pact
             3                           Uses pact
             4                           Uses pact
             5                         Uses fitbit
             6                         Uses fitbit
             7                         Uses fitbit
             8                         Uses fitbit
             9                         Uses fitbit
             10    Asked GF to ring in the morning
             11    Asked GF to ring in the morning
             12    Asked GF to ring in the morning
             13    Asked GF to ring in the morning
             14    Asked GF to ring in the morning
             15      Has read "The Power of Habit"
             16      Has read "The Power of Habit"
             17      Has read "The Power of Habit"
             18      Has read "The Power of Habit"
             19               Shares with his wife
             20               Shares with his wife
             21               Shares with his wife
             22               Shares with his wife
Pain         0                                   1
             1                                   1
             2                                   1
             3                                   1
             4                                   1
             5                                   2
             6                                   2
                                ...               
Code         16     (challenges, passive, reading)
             17     (challenges, passive, reading)
             18     (challenges, passive, reading)
             19           (accountability, wife, )
             20           (accountability, wife, )
             21           (accountability, wife, )
             22           (accountability, wife, )
Demographic  0               (job, entrepreneur, )
             1            (job, creative, blogger)
             2             (job, creative, writer)
             3                     (habit, diet, )
             4                 (habit, exercise, )
             5               (job, entrepreneur, )
             6            (job, creative, blogger)
             7             (job, creative, writer)
             8                     (habit, diet, )
             9                 (habit, exercise, )
             10              (job, entrepreneur, )
             11           (job, creative, blogger)
             12            (job, creative, writer)
             13                    (habit, diet, )
             14                (habit, exercise, )
             15                     (job, coach, )
             16              (job, entrepreneur, )
             17                 (habit, writing, )
             18              (habit, wakeing-up, )
             19                     (job, coach, )
             20              (job, entrepreneur, )
             21                 (habit, writing, )
             22              (habit, wakeing-up, )
dtype: object

In [146]:

    

prob_demo_pivot = prob_demo.pivot_table(index='Code', columns='Demographic')
prob_demo_pivot.index = pd.MultiIndex.from_tuples(prob_demo_pivot.index)
prob_demo_pivot


    

    
Out[146]:






  

    

      

      

      

      
Pain
    
    

      

      

      
Demographic
      
(habit, diet, )
      
(habit, exercise, )
      
(habit, wakeing-up, )
      
(habit, writing, )
      
(job, coach, )
      
(job, creative, blogger)
      
(job, creative, writer)
      
(job, entrepreneur, )
    
  
  

    

      
accountability
      
uses-product
      

      
1
      
1
      
NaN
      
NaN
      
NaN
      
1
      
1
      
1.0
    
    

      
wife
      

      
2
      
2
      
1
      
1
      
1
      
2
      
2
      
1.5
    
    

      
challenges
      
passive
      
reading
      
NaN
      
NaN
      
1
      
1
      
1
      
NaN
      
NaN
      
1.0
    
    

      
tracking
      
uses-product
      

      
2
      
2
      
NaN
      
NaN
      
NaN
      
2
      
2
      
2.0
    
  

In [ ]:

    




    

In [ ]: