Linear regression with machine learning (scikit-learn) Part 1

Testing the machine learning algorithm for regression, dataset populated with fake data.

Task:

You are owner of imaginary shop that sells stuff online via two channels: mobile app and through website. You are given dataset with customer info, based on such info we should decide whether it is better to invest more to enhancing the website or the mobile app to drive sales.

In this section we will inspect the data to get a feel for it, the next section will deal with machine learning.

In [9]:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline
In [10]:
# load fake customer data
customers = pd.read_csv('Ecommerce Customers')
customers.head()
Out[10]:
Email Address Avatar Avg. Session Length Time on App Time on Website Length of Membership Yearly Amount Spent
0 mstephenson@fernandez.com 835 Frank Tunnel\nWrightmouth, MI 82180-9605 Violet 34.497268 12.655651 39.577668 4.082621 587.951054
1 hduke@hotmail.com 4547 Archer Common\nDiazchester, CA 06566-8576 DarkGreen 31.926272 11.109461 37.268959 2.664034 392.204933
2 pallen@yahoo.com 24645 Valerie Unions Suite 582\nCobbborough, D... Bisque 33.000915 11.330278 37.110597 4.104543 487.547505
3 riverarebecca@gmail.com 1414 David Throughway\nPort Jason, OH 22070-1220 SaddleBrown 34.305557 13.717514 36.721283 3.120179 581.852344
4 mstephens@davidson-herman.com 14023 Rodriguez Passage\nPort Jacobville, PR 3... MediumAquaMarine 33.330673 12.795189 37.536653 4.446308 599.406092
In [3]:
customers.describe()
Out[3]:
Avg. Session Length Time on App Time on Website Length of Membership Yearly Amount Spent
count 500.000000 500.000000 500.000000 500.000000 500.000000
mean 33.053194 12.052488 37.060445 3.533462 499.314038
std 0.992563 0.994216 1.010489 0.999278 79.314782
min 29.532429 8.508152 33.913847 0.269901 256.670582
25% 32.341822 11.388153 36.349257 2.930450 445.038277
50% 33.082008 11.983231 37.069367 3.533975 498.887875
75% 33.711985 12.753850 37.716432 4.126502 549.313828
max 36.139662 15.126994 40.005182 6.922689 765.518462
In [4]:
customers.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 500 entries, 0 to 499
Data columns (total 8 columns):
Email                   500 non-null object
Address                 500 non-null object
Avatar                  500 non-null object
Avg. Session Length     500 non-null float64
Time on App             500 non-null float64
Time on Website         500 non-null float64
Length of Membership    500 non-null float64
Yearly Amount Spent     500 non-null float64
dtypes: float64(5), object(3)
memory usage: 31.3+ KB

Time spent on website and amount of money spent yearly do not correlate much

In [5]:
sns.jointplot(data=customers, x="Time on Website", y="Yearly Amount Spent", kind="hex")
Out[5]:
<seaborn.axisgrid.JointGrid at 0x7f9e63fc7ba8>

On the other hand the amount of time spent on the app correlates more to yearly revenues, so the more time is spent on app the higher revenues we have in our eshop.

In [6]:
sns.jointplot(data=customers, x="Time on App", y="Yearly Amount Spent", kind="hex")
Out[6]:
<seaborn.axisgrid.JointGrid at 0x7f9e63e6e390>

Let's see if there are any other interesting correlations. We will see that length of membership is strongly correlated to the amount that is spent yearly.

In [7]:
sns.pairplot(customers)
Out[7]:
<seaborn.axisgrid.PairGrid at 0x7f9e63a93a58>
In [8]:
sns.lmplot(data=customers, x="Length of Membership", y="Yearly Amount Spent")
Out[8]:
<seaborn.axisgrid.FacetGrid at 0x7f9e623c6fd0>

Linear fit is having rather narrow error range, indicating nice accuracy of the linear model.

Source:

code snippets from Jose Portilla Udemy course "Python for Data Science and Machine Learning Bootcamp" https://www.udemy.com/python-for-data-science-and-machine-learning-bootcamp/