https://github.com/okaditya84/ecommerce-transactions
Outlined in this piece is an evaluation of core corporate metrics derived from various data files.
https://github.com/okaditya84/ecommerce-transactions
analysis data-science eda lookalike python segmentation
Last synced: 3 months ago
JSON representation
Outlined in this piece is an evaluation of core corporate metrics derived from various data files.
- Host: GitHub
- URL: https://github.com/okaditya84/ecommerce-transactions
- Owner: okaditya84
- License: mit
- Created: 2025-01-27T16:41:40.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-01-27T18:28:30.000Z (over 1 year ago)
- Last Synced: 2025-02-05T21:05:24.304Z (over 1 year ago)
- Topics: analysis, data-science, eda, lookalike, python, segmentation
- Language: Jupyter Notebook
- Homepage:
- Size: 1.91 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# eCommerce Transaction Analysis
A project demonstrating an end-to-end workflow for exploring data, identifying lookalike customers, and segmenting customers using clustering techniques. Three Jupyter notebooks are included for each task.
---
## Table of Contents
1. [Overview](#overview)
2. [File Structure](#file-structure)
3. [Task Descriptions](#task-descriptions)
4. [Dependencies](#dependencies)
5. [Usage](#usage)
6. [License](#license)
---
## Overview
This repository analyzes an e-commerce dataset to uncover insights into customer behavior, purchase patterns, and possible marketing strategies.
---
## File Structure
- **Aditya_Jethani_EDA.ipynb** (Task 1): Exploratory Data Analysis
- **Aditya_Jethani_Lookalike.ipynb** (Task 2): Lookalike customer modeling
- **Aditya_Jethani_Clustering.ipynb** (Task 3): K-Means clustering for customer segmentation
- **Customers.csv, Products.csv, Transactions.csv**: Example input data files
- **README.md**: Project documentation
- **LICENSE**: MIT License
---
## Task Descriptions
### Task 1: Exploratory Data Analysis (EDA)
- Merges multiple datasets to create a complete view of customers, products, and transactions.
- Examines sales trends, category-wise revenue, and other key metrics.
- Visualizes daily and monthly revenue, top-selling items, and customer segments.
### Task 2: Lookalike Modeling
- Computes features (e.g., total spend, frequency) for a set of target customers.
- Uses cosine similarity to find the closest matches (lookalikes) based on transactional or demographic profiles.
- Outputs top matches and saves them to a CSV for further use.
### Task 3: Customer Segmentation (K-Means Clustering)
- Scales relevant customer metrics with `StandardScaler`.
- Determines the optimal number of clusters using metrics like Silhouette Score and Davies-Bouldin Index.
- Assigns cluster labels, visualizes the results, and interprets cluster profiles.
---
## Dependencies
Install the following Python libraries:
```bash
pandas
numpy
matplotlib
seaborn
scikit-learn