Skip to content

This model is capable of categorizing customers into distinct groups based on common characteristics, enabling personalized marketing strategies and improving customer satisfaction and retention.

Notifications You must be signed in to change notification settings

tejatanush/Customer-Segmentation-System

Repository files navigation

Customer-Segmentation-System

This model is capable of categorizing customers into distinct groups based on common characteristics, enabling personalized marketing strategies and improving customer satisfaction and retention.

Features

Import required libraries
Import dataset
Data Preprocessing
Find and fill missing values
Encoding data
Feature Scaling
Dimensionality Reduction
Build a model
K-Mean Clustering
Finding Optimal number of clusters
Build a K-Mean clustering model
Visualizing Clusters
Observing percentage of customers in each cluster
Evaluate K-Means cluster model
Silhouette score
Davies bouldin score
Hierarchical Clustering
Finding Optimal number of clusters
Build a Hierarchical clustering model
Visualizing Clusters
Observing percentage of customers in each cluster
Evaluate Hierarchical cluster model
Silhouette score
Davies bouldin score
Finalize the model
Cluster Analysis
Profiling

Applications

Personalized Marketing: Tailoring marketing campaigns and promotions to specific customer segments based on their preferences and behavior.
Product Development: Informing product design and feature enhancements by understanding the needs and desires of different customer groups.
Customer Retention Strategies: Crafting targeted retention programs and loyalty rewards to increase customer loyalty within specific segments.
Pricing Optimization: Adjusting pricing strategies for different segments to maximize revenue while meeting the price sensitivity and expectations of each group.

Compiler Type:

I used a compiler CPU provided in google colab while running this project.

Results

This project contain two models. Among them K-Means clustering model have a davies bouldin score of 1.14 and silhouette score of 0.283.....Hierarchical clustering model have a davies bouldin score of 1.613 and silhouette score of 0.229. So K-Means model performed well with 4 clusters.

These are the 4 clusters through K-Means

CLUSTER-0: IMPULSIVE SPENDERS

This cluster is having the customers with moderate age,moderate income and high spending score. It mean this cluster contains the customers who are in youth age and having moderate annual income but spending lot's of money.

CLUSTER-1: WEALTHY ELDERS

This cluster is having high age, high annual income and high spending score. It mean this cluster contain customers who are very old aged and having high annual income and also spending more money daily.

CLUSTER-2: AFFLUENT MODERATES

This cluster is having moderate age, high annual income and moderate spending score.It mean this cluster contain customers who are middle aged, having high annual income and spending moderate money as per their needs in a planned way.

CLUSTER-3: STABLE RETIREES

This cluster is having high age, moderate annual income and moderate spending score. It mean this cluster contain customers who are old aged, having moderate income and moderate spending as per their needs.

References

Customer_Segmentation_System.csv-> (https://www.kaggle.com/datasets/govindkrishnadas/segment)

Requirements

Pandas
Numpy
Matplotlib
sci-kit learn
scipy

Programmes

Python

Owner

Teja Tanush

About

This model is capable of categorizing customers into distinct groups based on common characteristics, enabling personalized marketing strategies and improving customer satisfaction and retention.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published