K-Means Clustering

A Short Case Study

On this notebook, we're gonna dive not so deep on the basics of how to dow a K-Means Clustering on a small example dataset. At the end of
this study, I hope we could achieve the following understandings regarding our problem:

1. What's a good way to segment our dataset on a small set of clusters?

2. How can we achieve quick results using the Pandas, Numpy, Matplotlib, Pyplot and SKLearn modules?
3. How the select the best hyperparameters for K-Means Clustering?
4. How to display and visualize data in the most honest and friendly way to our stakeholders?

Initialization

# Initialization

# Module Imports
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
from sklearn.cluster import KMeans

# Style Definitions
plt.style.use('Solarize_Light2')

# Get dataframe from CSV file

df = pd.read_csv('customers.csv')
df.head()

CustomerID Gender Age Annual Income (k$) Spending Score (1-100)

0 1 Male 19 15 39

1 2 Male 21 15 81

2 3 Female 20 16 6

3 4 Female 23 16 77

4 5 Female 31 17 40

df.shape

(200, 5)

df.describe()

CustomerID Age Annual Income (k$) Spending Score (1-100)

count 200.000000 200.000000 200.000000 200.000000

mean 100.500000 38.850000 60.560000 50.200000

std 57.879185 13.969007 26.264721 25.823522

min 1.000000 18.000000 15.000000 1.000000

25% 50.750000 28.750000 41.500000 34.750000

50% 100.500000 36.000000 61.500000 50.000000

75% 150.250000 49.000000 78.000000 73.000000

max 200.000000 70.000000 137.000000 99.000000

plt.figure(1, figsize=(16,4))
n = 0
for i in ['Age', 'Annual Income (k$)', 'Spending Score (1-100)']:
n += 1
 plt.subplot(1 , 3 , n)
plt.subplots_adjust(hspace =0.5 , wspace = 0.5)
sns.distplot(df[i] , bins = 32)
plt.title(f'Histogram of {i}')
plt.show()

# Assignment Stage

X1 = df.loc[:, ['Age', 'Spending Score (1-100)']].values

inertia = []
for n in range(1 , 11):
model = KMeans(n_clusters = n,
init='k-means++',
max_iter=500,
random_state=42)
model.fit(X1)
inertia.append(model.inertia_)

plt.figure(1 , figsize = (15 ,6))

plt.plot(np.arange(1 , 11) , inertia , 'o')
plt.plot(np.arange(1 , 11) , inertia , '-' , alpha = 0.5)
plt.xlabel('Number of Clusters') , plt.ylabel('Inertia')
plt.show()

model = KMeans(n_clusters = 4,
init='k-means++',
max_iter=500,
random_state=42)
model.fit(X1)
labels = model.labels_
centroids = model.cluster_centers_
y_kmeans = model.fit_predict(X1)

plt.figure(figsize=(20,10))
plt.scatter(X1[y_kmeans == 0, 0], X1[y_kmeans == 0, 1], s = 100, c = 'red', label = 'Cluster 1')
plt.scatter(X1[y_kmeans == 1, 0], X1[y_kmeans == 1, 1], s = 100, c = 'blue', label = 'Cluster 2')
plt.scatter(X1[y_kmeans == 2, 0], X1[y_kmeans == 2, 1], s = 100, c = 'green', label = 'Cluster 3')
plt.scatter(X1[y_kmeans == 3, 0], X1[y_kmeans == 3, 1], s = 100, c = 'cyan', label = 'Cluster 4')
plt.scatter(X1[y_kmeans == 4, 0], X1[y_kmeans == 4, 1], s = 100, c = 'magenta', label = 'Cluster 5')
plt.title('Clusters of Customers - Age X Spending Score')
plt.xlabel('Age')
plt.ylabel('Spending Score')
plt.legend()
plt.show()

Second Clustering
By Annual Income and Spending Score

# Assignment Stage

X2 = df.loc[:, ['Annual Income (k$)', 'Spending Score (1-100)']].values

inertia = []
for n in range(1 , 11):
model = KMeans(n_clusters = n,
init='k-means++',
max_iter=500,
random_state=42)
model.fit(X2)
inertia.append(model.inertia_)

plt.figure(1 , figsize = (20, 10))

plt.plot(np.arange(1 , 11) , inertia , 'o')
plt.plot(np.arange(1 , 11) , inertia , '-' , alpha = 0.5)
plt.xlabel('Number of Clusters') , plt.ylabel('Inertia')
plt.show()
model = KMeans(n_clusters = 5,
init='k-means++',
max_iter=500,
random_state=42)
model.fit(X2)
labels = model.labels_
centroids = model.cluster_centers_
y_kmeans = model.fit_predict(X2)

plt.figure(figsize=(20,10))
plt.scatter(X2[y_kmeans == 0, 0], X2[y_kmeans == 0, 1], s = 100, c = 'red', label = 'Cluster 1')
plt.scatter(X2[y_kmeans == 1, 0], X2[y_kmeans == 1, 1], s = 100, c = 'blue', label = 'Cluster 2')
plt.scatter(X2[y_kmeans == 2, 0], X2[y_kmeans == 2, 1], s = 100, c = 'green', label = 'Cluster 3')
plt.scatter(X2[y_kmeans == 3, 0], X2[y_kmeans == 3, 1], s = 100, c = 'cyan', label = 'Cluster 4')
plt.scatter(X2[y_kmeans == 4, 0], X2[y_kmeans == 4, 1], s = 100, c = 'magenta', label = 'Cluster 5')
plt.title('Clusters of Customers - Annual Income (k$) X Spending Score')
plt.xlabel('Annual Income (k$)')
plt.ylabel('Spending Score')
plt.legend()
plt.show()

# Assignment Stage

from sklearn.cluster import KMeans

X3 = df.loc[:, ['Age', 'Annual Income (k$)', 'Spending Score (1-100)']].values

inertia = []
for n in range(1 , 11):
model = KMeans(n_clusters = n,
init='k-means++',
max_iter=500,
random_state=42)
model.fit(X3)
inertia.append(model.inertia_)

plt.figure(1 , figsize = (20, 10))

plt.plot(np.arange(1 , 11) , inertia , 'o')
plt.plot(np.arange(1 , 11) , inertia , '-' , alpha = 0.5)
plt.xlabel('Number of Clusters') , plt.ylabel('Inertia')
plt.show()
 model = KMeans(n_clusters = 6,
init='k-means++',
max_iter=500,
random_state=42)
model.fit(X3)
labels = model.labels_
#centroids = model.cluster_centers_

df['cluster'] = labels
df

CustomerID Gender Age Annual Income (k$) Spending Score (1-100) cluster

0 1 Male 19 15 39 0

1 2 Male 21 15 81 1

2 3 Female 20 16 6 0

3 4 Female 23 16 77 1

4 5 Female 31 17 40 0

... ... ... ... ... ... ...

195 196 Female 35 120 79 4

196 197 Female 45 126 28 2

197 198 Male 32 126 74 4

198 199 Male 32 137 18 2

199 200 Male 30 137 83 4

200 rows × 6 columns

fig = px.scatter_3d(df,
x="Age",
y="Annual Income (k$)",
z="Spending Score (1-100)",
color='cluster',
hover_data=["Age",
"Annual Income (k$)",
"Spending Score (1-100)"],
category_orders = {"cluster": range(0, 5)},
)

fig.update_layout(margin=dict(l=0, r=0, b=0, t=0))

fig.show()