This project analyzes a retail dataset to segment customers based on their purchasing behavior using clustering techniques and outlier detection. The goal is to identify distinct customer segments, such as high spenders, frequent buyers, and valuable outliers, so that targeted marketing strategies can be developed.
-
Data Cleaning & Preprocessing:
The notebook performs data cleaning and feature engineering, such as calculating total spend, frequency of purchases, and recency. -
Aggregation & Outlier Detection:
Customers are aggregated based on key metrics, and outliers are identified using the Interquartile Range (IQR) method. These outliers are then separated for further analysis. -
Clustering & Segmentation:
A clustering algorithm (e.g., K-means) is applied on the engineered features to group customers into segments. Each cluster is then labeled with a short rationale describing its characteristics. -
Visualization:
Various visualizations, including scatter plots and pair plots, are generated to visualize the clusters and understand the customer distribution.
- Python 3.x
- Jupyter Notebook
- Libraries:
pandas,numpy,matplotlib,seaborn,scikit-learn,openpyxl
You can install the necessary libraries using pip:
pip install -r requirements.txt