Segment Sphere: RFM Customer Segmentation

An interactive customer segmentation platform using RFM analysis to provide actionable insights on customer behavior.

---

Project Description

Segment Sphere is a tool designed to analyze customer data and provide actionable insights through RFM analysis (Recency, Frequency, and Monetary). It enables businesses to segment their customers into meaningful groups, helping them better understand customer behavior and improve decision-making.

---

Table of Contents

• Project Title and Overview
• Project Description
• Features
• Technology Stack
• Preview
• Installation
• Usage
• Configuration
• Testing
• License
• FAQs
• Contact Information

---

Features

• Automated RFM analysis for customer segmentation.
• Interactive visualizations using Plotly for actionable insights.
• User-friendly web interface for data upload and visualization.
• Scalable architecture supporting large datasets.

---

Technology Stack

Frontend:

• HTML5, CSS3, JavaScript (ES6)
• Bootstrap for responsive design

Backend:

• Python, Flask (for RESTful API)
• Pandas, NumPy (for data preprocessing)
• Scikit-learn (for K-means clustering)

Other Tools:

• Plotly (for chart visualizations)

---

Preview

Below are previews showcasing a selection of insightful visualizations generated by Segment Sphere. Explore the platform to dive deeper into more detailed charts and comprehensive RFM analysis insights!

---

Installation

To set up the project locally, follow these steps:

1. Clone the repository:

   git clone https://github.com/ThakkarVidhi/rfm-customer-segmentation.git
   cd rfm-customer-segmentation

2. Install the necessary dependencies:

   pip install -r requirements.txt

3. Set up the database and run the app:

   python main.py

4. Open http://127.0.0.1:5000/ in your browser to start using the application.

---

Usage

To run the RFM analysis, follow these steps:

1. Upload your CSV file containing customer data (with columns: InvoiceNo, StockCode, Description, Quantity, InvoiceDate, UnitPrice, CustomerID, Country).
2. Select the appropriate file and click "Analyze".
3. View the segmentation results and explore the interactive charts.

---

Configuration

This project uses a config.py file for managing application settings and paths. The configuration file contains the following settings:

Configuration Settings:

• BASE_DIR: The absolute path to the base directory of the project.
• STATIC_DIR: Path to the static directory used to serve static files (e.g., CSS, JavaScript, images).
• INPUT_DATA_DIR: Path to the directory where input data (e.g., CSV files) is stored.
• TEMPLATE_DIR: Path to the directory containing HTML templates for the application.
• MODEL_PATH: Path to the saved machine learning model used for customer segmentation.
• DEBUG: Set to True for enabling debugging during development.

---

Testing

This project does not include automated tests. To test the application, simply open it in your browser and interact with the site.

---

License

This project is licensed under the MIT License.

---

FAQs

Q: What data format is required for upload?
A: CSV files with columns for InvoiceNo, StockCode, Description, Quantity, InvoiceDate, UnitPrice, CustomerID, and Country.

Q: How can I modify the application’s configuration?
A: The configuration can be modified by editing the config.py file. You can change the paths and other settings like DEBUG.

Q: What should I do if the application is not displaying correctly?
A: Make sure the required directories and files are in place. Check the browser's developer tools for errors, and verify that the paths in config.py are correct.

Q: Can I use a different machine learning model for segmentation?
A: Yes, you can replace the model file (customer_segmentation_model.pkl) in the app/models directory with another compatible model as long as it follows the same format.

---

Contact Information

For questions or feedback, you can reach me at vidhithakkar.ca@gmail.com or via LinkedIn.