amazon-recommendation-system

• SQL
• Python (pandas, numpy, scikit-learn, scipy, statsmodels)
• Google Colab Built using Python, SQL, and Google Colab for model development and experimentation.

Run in Google Colab

🚀 Quick Start

Clone the repository

git clone https://github.com/YOUR_USERNAME/amazon-recommendation-system.git
cd amazon-recommendation-system



## How to Run

This project is designed to run in Google Colab.

### Option 1: Open directly in Colab
Click the Colab badge above.

### Option 2: Run manually in Colab
In a new Colab notebook, run:

```python
!git clone https://github.com/Nejatbakhsh-y/amazon-recommendation-system.git
%cd amazon-recommendation-system
!pip install -r requirements.txt



# Amazon-Style Recommendation System with A/B Testing

## Project Overview
This project builds a recommendation system using collaborative filtering on user-product interaction data.

## Dataset
Synthetic dataset including views, clicks, and purchases.

## Methodology
- SQL: interaction aggregation, train/test split
- Python: SVD-based collaborative filtering
- Evaluation: CTR, Conversion Rate

## Results
| Metric | Control | Treatment |
|--------|--------:|----------:|
| CTR | 8.0% | 9.5% |
| Conversion Rate | 15.0% | 16.0% |

## Tech Stack
- Python (pandas, numpy, scikit-learn, scipy, statsmodels)
- SQL
- Google Colab (for development and execution)

## How to Run
Open the Colab notebook in:
notebooks/recommendation_model.ipynb

## Results

| Metric | Control | Treatment |
|--------|--------:|----------:|
| CTR | 8.0% | 9.5% |
| Conversion Rate | 15.0% | 16.0% |

The collaborative filtering model improved both engagement and post-click conversion performance.