Universal CSV Analytics Platform

#live demo https://universal-csv-analytics.vercel.app/

A professional, dynamic Data Analytics Platform that can ingest any CSV file, automatically analyze its structure, map its columns, and generate a comprehensive interactive dashboard containing KPIs, Plotly charts, AI Business Insights, and downloadable reports.

Built as a 2nd-year Computer Science major project for a Data Analytics Bootcamp.

✨ Features

• Smart Dataset Detection: Automatically identifies standard metrics like Sales, Profit, Customer, Product, Region, and Date based on common synonyms.
• Adaptive Dashboard: Modifies the available charts and insights dynamically. If a dataset doesn't have "Sales", it gracefully falls back to generic statistical analysis (correlations, distributions) without crashing.
• Interactive Visualizations: Powered by Plotly, featuring tooltips, zoom, and dynamic rendering.
• AI Business Insights: Automatically generates rule-based business insights and recommendations (e.g., Identifying the most profitable category or loss-making products).
• Data Cleaning Engine: Automatically handles missing values and drops exact duplicates.
• Downloadable Reports: Export the cleaned data to CSV/Excel, or generate a professional PDF Report summarizing the dataset and insights using ReportLab.
• Modern UI: Glassmorphism design, gradient text, smooth animations, drag-and-drop file upload, and Dark/Light mode toggle.

🛠️ Tech Stack

• Backend: Python, Flask
• Data Processing: Pandas, NumPy
• Visualizations: Plotly, Matplotlib (fallback/generic), Seaborn (fallback/generic)
• Report Generation: ReportLab (PDF), OpenPyXL (Excel)
• Frontend: HTML5, CSS3 (Vanilla), JavaScript (Vanilla), Bootstrap Icons

📂 Project Structure

Universal-CSV-Analytics/
│── uploads/                 # Temporary storage for uploaded CSVs
│── reports/                 # Temporary storage for generated PDFs/Excel files
│── templates/
│     └── index.html         # Main Single Page Application UI
│── static/
│     ├── style.css          # Styling & Themes (Dark/Light mode)
│     └── script.js          # API interactions & Chart rendering
│── main.py                  # Flask Application, Routing & Data Cleaning
│── charts.py                # Plotly Chart generation logic
│── requirements.txt         # Project Dependencies
│── README.md                # Documentation

🚀 Setup & Installation

1. Clone the repository:

   git clone <repo_url>
   cd Universal-CSV-Analytics

2. Create a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install dependencies:

   pip install -r requirements.txt

4. Run the application:

   python main.py

5. Access the application: Open your browser and navigate to http://127.0.0.1:5000/

📊 How It Works

1. Upload: Drag and drop any .csv file into the upload zone.
2. Process: The backend reads the CSV using Pandas, imputes missing values, and calculates a Data Health Score.
3. Map: The system looks for column synonyms (e.g., mapping "Revenue" to "Sales").
4. Analyze: The backend generates KPIs and Plotly JSON structures via charts.py.
5. Render: The frontend consumes the JSON and uses Plotly.js to render beautiful interactive charts.
6. Export: Click the download buttons to generate a Python ReportLab PDF or download cleaned data.

🤝 Contributing

Contributions, issues, and feature requests are welcome!

📝 License

This project is open-source and available under the MIT License.