Scribd
Upload
134 views3 pages

Salesforce PD1

The Python code performs K-Means clustering on a customer dataset to determine the optimal number of clusters. It loads customer data, checks for null values, creates a scatter plot of Age vs Spending Score, uses the elbow method to find the best k between 1-10, and plots the clustered data points. It also shows an alternative sub-optimal approach that manually selects k=6.

Uploaded by

Vigneshwaran Ganapathi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
134 views3 pages

Salesforce PD1

The Python code performs K-Means clustering on a customer dataset to determine the optimal number of clusters. It loads customer data, checks for null values, creates a scatter plot of Age vs Spending Score, uses the elbow method to find the best k between 1-10, and plots the clustered data points. It also shows an alternative sub-optimal approach that manually selects k=6.

Uploaded by

Vigneshwaran Ganapathi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 3

Que : You work in XYZ Company as a Python Developer.

The company officials want you to


write code for a clustering problem. Dataset: customers.csv Tasks to be performed: 1. K-
Means Clustering: - Load customer data. - Check the number of cells in each column with
null values. - Create a scatter plot with Age as X and Spending Score as Y. - Find out the best
number for clusters between 1 and 10 (inclusive) using the elbowmethod. - Draw a scatter
plot displaying data points colored on the basis of clusters

For Optimal:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from yellowbrick.cluster import KElbowVisualizer

# Load customer data


data = pd.read_csv('customers.csv')

# Check for null values


print(data.isnull().sum())

# Create a scatter plot with Age as X and Spending Score as Y


plt.scatter(data['Age'], data['Spending Score (1-100)'])
plt.xlabel('Age')
plt.ylabel('Spending Score (1-100)')
plt.show()

# Use the optimal k-means clustering algorithm to determine the number of clusters
model = KMeans()
visualizer = KElbowVisualizer(model, k=(1,10))
visualizer.fit(data[['Age', 'Spending Score (1-100)']])
visualizer.show()

# Draw a scatter plot displaying data points colored on the basis of clusters
optimal_k = visualizer.elbow_value_
kmeans = KMeans(n_clusters=optimal_k, init='k-means++', max_iter=300, n_init=10,
random_state=0)
clusters = kmeans.fit_predict(data[['Age', 'Spending Score (1-100)']])
data['Cluster'] = clusters
plt.scatter(data['Age'], data['Spending Score (1-100)'], c=data['Cluster'], cmap='viridis')
plt.xlabel('Age')
plt.ylabel('Spending Score (1-100)')
plt.show()

For Sub-Optimal:
import pandas as pd

import numpy as np

import matplotlib.pyplot as plt

from sklearn.cluster import KMeans

# Load customer data

data = pd.read_csv('customers.csv')

# Check for null values

print(data.isnull().sum())

# Create a scatter plot with Age as X and Spending Score as Y

plt.scatter(data['Age'], data['Spending Score (1-100)'])

plt.xlabel('Age')
plt.ylabel('Spending Score (1-100)')

plt.show()

# Use the suboptimal k-means clustering algorithm to determine the number of clusters

sum_of_squared_distances = []

K = range(1,11)

for k in K:

km = KMeans(n_clusters=k, init='k-means++', max_iter=300, n_init=10, random_state=0)

km = km.fit(data[['Age', 'Spending Score (1-100)']])

sum_of_squared_distances.append(km.inertia_)

plt.plot(K, sum_of_squared_distances, 'bx-')

plt.xlabel('Number of Clusters')

plt.ylabel('Sum of Squared Distances')

plt.title('Elbow Method For Optimal k')

plt.show()

# Draw a scatter plot displaying data points colored on the basis of clusters

suboptimal_k = 6 # a value selected as an example

kmeans = KMeans(n_clusters=suboptimal_k, init='k-means++', max_iter=300, n_init=10,


random_state=0)

clusters = kmeans.fit_predict(data[['Age', 'Spending Score (1-100)']])

data['Cluster'] = clusters

plt.scatter(data['Age'], data['Spending Score (1-100)'], c=data['Cluster'], cmap='viridis')

plt.xlabel('Age')

plt.ylabel('Spending Score (1-100)')

plt.show()

You might also like