https://github.com/sreyash1mohanty/e-commerce_marketing_machine_learning

Machine Learning in Marketing using various machine learning algorithms.
https://github.com/sreyash1mohanty/e-commerce_marketing_machine_learning

churn-prediction customer decision-trees knn logistic-regression machine-learning machine-learning-algorithms segmentation

Last synced: 2 months ago
JSON representation

Machine Learning in Marketing using various machine learning algorithms.

• Host: GitHub
• URL: https://github.com/sreyash1mohanty/e-commerce_marketing_machine_learning
• Owner: sreyash1mohanty
• Created: 2025-03-21T17:23:00.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2025-03-25T13:25:53.000Z (7 months ago)
• Last Synced: 2025-03-29T10:18:26.895Z (7 months ago)
• Topics: churn-prediction, customer, decision-trees, knn, logistic-regression, machine-learning, machine-learning-algorithms, segmentation
• Language: Jupyter Notebook
• Homepage:
• Size: 4.34 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# E-COMMERCE_MARKETING_MACHINE_LEARNING

 The goal of marketing in e-commerce is to attract, engage, and retain customers through targeted campaigns, improving customer lifetime value ,CLV and brand loyalty..

# Customer Segmentation & Churn Prediction  

## Overview  

This repository contains two machine learning projects focused on customer analytics:  

1. **Customer Segmentation using RFM Analysis & K-Means Clustering**  

2. **Customer Churn Prediction in an E-commerce Dataset**  

Both projects involve data preprocessing, feature engineering, model training, and evaluation to extract meaningful insights and improve business decision-making.  

---  

## Project 1: Customer Segmentation (UK-Based Online Retail Dataset)  

### Objective  

Customer segmentation is crucial in modern customer-centric marketing. This project uses **RFM analysis (Recency, Frequency, Monetary value)** and **K-Means Clustering** to segment customers based on purchasing behavior.  

### Dataset  

- The dataset is from an online UK-based retail store, covering transactions from **01/12/2009 to 09/12/2010**.  

- The company mainly sells unique, all-occasion giftware.  

- Many customers are wholesalers.  

### Methodology  

1. **Data Preprocessing**:  

   - Handled missing values and outliers.  

   - Filtered relevant transactions.  

2. **Feature Engineering (RFM Analysis)**:  

   - **Recency**: How recently a customer made a purchase.  

   - **Frequency**: How often a customer purchases.  

   - **Monetary Value**: Total spending of a customer.  

3. **Standardization**:  

   - Scaled the RFM values using **MinMaxScaling**.  

4. **Clustering using K-Means**:  

   - Chose **3 clusters** for segmentation.  

   - Identified customer groups based on spending patterns.  

5. **Prediction for Incoming Customers**:  

   - Trained a model to predict the segment of new customers.  

6. **Modeling Tools Used**:  

   - **TensorFlow** for training the predictive model.  

   - **Scikit-Learn** for clustering.  

---

## Project 2: Customer Churn Prediction (E-commerce Dataset)  

### Objective  

This project aims to predict whether a customer will churn using an **e-commerce dataset** with over **5,000 records** and **20 features**.  

### Methodology  

1. **Data Cleaning & Preprocessing**:  

   - Handled missing values, duplicate entries, and outliers.  

   - Standardized features for consistency.  

2. **Feature Selection**:  

   - Analyzed relationships between **churn** and different attributes.  

   - Kept optimal features to improve accuracy.  

3. **Model Training & Evaluation**:  

   - **Trained multiple models** to determine the best one:  

     - Logistic Regression  

     - Decision Trees  

     - Random Forests  

     - Support Vector Machine (SVM)  

   - **Best Model**: Random Forest with **94% accuracy**.  

   - **Evaluation Metrics**:  

     - Confusion Matrix  

     - Precision, Recall, F1-score  

---

## Results & Insights  

- **Customer Segmentation:**  

  - Successfully categorized customers into 3 segments based on purchasing behavior.  

  - Helps businesses target different groups with personalized marketing strategies.  

- **Churn Prediction:**  

  - Achieved **94% accuracy** in predicting customer churn using **Random Forest**.  

  - Provides insights into the key factors contributing to customer churn.  

---

## Technologies Used

- Python

- Pandas, NumPy, Matplotlib, Seaborn (Data Analysis & Visualization)

- Scikit-Learn (Machine Learning Models)

- TensorFlow (Predictive Model for Customer Segmentation)