Dendrogram visualization in SciPy - Mini Project: Build & Apply

Dendrogram visualization

📖 Scenario: You work in a small company that wants to group similar products based on their features. You have collected data about the products and want to see how they cluster together visually.

🎯 Goal: Build a dendrogram visualization using scipy to show how products group based on their features.

📋 What You'll Learn

Create a data dictionary with product names and their feature values

Set a linkage method variable for clustering

Use scipy.cluster.hierarchy.linkage to compute clusters

Use scipy.cluster.hierarchy.dendrogram to plot the dendrogram

Print the dendrogram plot

💡 Why This Matters

🌍 Real World

Dendrograms help businesses group similar items or customers to understand patterns and make decisions.

💼 Career

Data scientists and analysts use dendrograms to visualize hierarchical clusters in marketing, biology, and many other fields.

1

Create product features data

Create a dictionary called products with these exact entries: 'ProductA': [1.0, 2.0], 'ProductB': [1.5, 1.8], 'ProductC': [5.0, 8.0], 'ProductD': [6.0, 9.0]

SciPy

# Create the products dictionary with exact keys and values
# Your code here

Need a hint?

Use curly braces to create a dictionary. Each product name is a key, and its features are a list of two numbers.

2

Set linkage method

Create a variable called linkage_method and set it to the string 'ward' to specify the clustering method.

SciPy

products = {
    'ProductA': [1.0, 2.0],
    'ProductB': [1.5, 1.8],
    'ProductC': [5.0, 8.0],
    'ProductD': [6.0, 9.0]
}
# Set the linkage_method variable to 'ward'
# Your code here

Need a hint?

Assign the string 'ward' to the variable linkage_method.

3

Compute linkage matrix

Import linkage from scipy.cluster.hierarchy. Create a list called data_points containing the feature lists from products. Use linkage(data_points, method=linkage_method) and save the result in a variable called Z.

SciPy

products = {
    'ProductA': [1.0, 2.0],
    'ProductB': [1.5, 1.8],
    'ProductC': [5.0, 8.0],
    'ProductD': [6.0, 9.0]
}
linkage_method = 'ward'
# Import linkage and create data_points list
# Compute linkage matrix Z
# Your code here

Need a hint?

Use a list comprehension to get the feature lists from the dictionary values.

4

Plot dendrogram

Import dendrogram from scipy.cluster.hierarchy and matplotlib.pyplot as plt. Use dendrogram(Z, labels=list(products.keys())) to create the dendrogram plot. Then call plt.show() to display it.

SciPy

from scipy.cluster.hierarchy import linkage

products = {
    'ProductA': [1.0, 2.0],
    'ProductB': [1.5, 1.8],
    'ProductC': [5.0, 8.0],
    'ProductD': [6.0, 9.0]
}
linkage_method = 'ward'

data_points = [features for features in products.values()]
Z = linkage(data_points, method=linkage_method)
# Import dendrogram and matplotlib.pyplot as plt
# Plot dendrogram with labels from products keys
# Show the plot
# Your code here

Need a hint?

Use dendrogram with Z and labels=list(products.keys()). Then call plt.show() to display the plot.