2. Introduction to Customer Segmentation
Customer segmentation divides customers into groups by
shared traits.
Enhances understanding of customer needs and targeting.
Uses Unsupervised Learning: KMeans Clustering to find
patterns automatically.
3. Background & Motivation
Limitations of Manual Segmentation
Inefficient for large datasets
Value of AI
Discovers hidden patterns in big data
Clustering Benefits
Groups customers without predefined labels
Business Impact
Personalised marketing and services
4. Project Objectives
Segment Customers
Using KMeans Clustering
Group on Key Features
Income, Age, Spending behaviour
Visualize Clusters
Via Principal Component Analysis (PCA)
Business Insights
Analyse cluster profiles
5. Dataset Overview
• Source: Kaggle - Marketing Campaign
• Demographics: Age, Education, Marital Status
• Spending on Wine, Fruits, Meat, etc.
• Total_Spend engineered by summing expenditures
6. Methodology
Preprocessing
• Drop unneeded columns
• Handle missing values
• Encode categorical data
Feature Selection
Choose Income, Recency, Education, Total Spend
KMeans Clustering
Apply 3 clusters, assign labels
Visualization
PCA to 2D, plot clusters
7. Code Highlights
Label Encoding
For categorical variables
StandardScaler
Data scaling for clustering
KMeans Model
n_clusters set to 3
PCA
Two-dimensional visualization
Cluster Analysis
Grouped data to profile clusters
8. Output Visualization
Scatter plot showing distinct clusters found by KMeans
Summary statistics table showing cluster characteristics (age, income,
spend)
9. Conclusion and Future Work
Successful
Segmentation
Three distinct clusters
identified
PCA Visualization
Clear cluster separation
observed
Improvement Ideas
Explore hierarchical clustering & DBSCAN
