🛒 AI Smart E-Commerce Analyzer

📌 Project Overview

This project analyzes e-commerce data to extract business insights and build a recommendation system using SQL and Python.

The project uses the Olist E-commerce dataset to understand customer behavior, product demand, revenue trends, and category relationships.

---

⚙️ Tech Stack

- Python (Pandas, Matplotlib)

- MySQL

- SQL

- Jupyter Notebook

- Git & GitHub

---

📂 Project Structure

AI_Smart_Ecommerce_Analyzer/

│

├── data/

│ ├── 00_raw/

│ └── 01_processed/

│

├── notebooks/

│ └── eda_analysis.ipynb

│

├── sql/

│ └── analysis.sql

│

├── src/

│

├── README.md

├── requirements.txt

└── .gitignore

---

📊 Current Progress

✅ Data Preparation

- Imported Olist dataset into MySQL

- Created database schema

- Structured tables:

- customers

- orders

- order_items

- products

---

✅ SQL Analysis

Performed SQL-based business analysis including:

- Top-selling products

- Top product categories

- Revenue analysis

- Customer regional analysis

- Repeat customer identification

- Customer retention metrics

- Monthly order trend analysis

---

✅ Recommendation System (SQL-based)

Built a basic recommendation engine using co-purchase analysis.

Features:

- Identified products frequently bought together

- Category-to-category recommendation logic

- Confidence-based recommendation approach

- Cross-category association analysis

---

✅ Exploratory Data Analysis (EDA)

Performed Python-based EDA using Jupyter Notebook.

Completed tasks:

- Dataset inspection

- Data type validation

- Missing value analysis

- Duplicate record checking

- Datetime conversion

- Monthly order trend visualization

- Product category distribution analysis

- Revenue distribution analysis

---

📈 Key Insights

- Certain product categories dominate sales volume

- Customer purchases show seasonal patterns

- Revenue distribution is highly skewed

- Repeat customers contribute significantly to total orders

- Cross-category buying patterns support recommendation logic

---

🚀 Next Steps

- Advanced EDA insights

- Customer segmentation using Python

- Dashboard creation

- Recommendation model enhancement

- Business KPI visualization

---

🎯 Project Goal

Build an AI-powered e-commerce analytics system capable of:

- Understanding customer purchase behavior

- Identifying business trends

- Generating product recommendations

- Supporting business decision-making through analytics

---

📌 Dataset

Dataset Used:

- Olist Brazilian E-commerce Dataset

Dataset contains:

- Customers

- Orders

- Products

- Payments

- Reviews

- Geolocation

- Sellers

---

📷 Future Additions

- Dashboard screenshots

- Recommendation output screenshots

- Trend analysis charts

- Business KPI visuals

---

👩‍💻 Author

Varshini

Aspiring Data Scientist | SQL | Python | Analytics | Recommendation Systems