Tapestra: The Concept Graph

An interactive Streamlit web app for visualizing and analyzing conceptual relationships within and across paragraphs of text. Tapestra uses NLP, graph theory, and physics-based visualization to help users explore the structure and flow of ideas in their writing.

---

🚀 Live App

Try the live app here!

---

📸 Screenshots

---

📝 Features

• Paragraph-by-Paragraph Concept Graphs:

  • Enter or edit paragraphs and generate directed concept graphs based on sentence similarity and importance.
  • Interactive physics controls (spring length, spring constant, gravity, central gravity) for each paragraph’s graph.

• Combined Concept Graph:

  • Combine all key nodes from all paragraphs into a single graph.
  • Separate physics controls for the combined graph.
  • Download any graph as an HTML file.

• Optimal Reading Path:

  • Table showing optimal reading sequences based on graph structure.

• Modern UI:

  • Responsive, wide layout with custom CSS.
  • All controls and graphs are always visible and interactive.

• Dark Theme:

  • Clean, dark-themed interface for readability and focus.

---

🖥️ Running Locally

Requirements

• Python 3.8 or higher
• See requirements.txt for dependencies

Install Dependencies

pip install -r requirements.txt

Run the App

streamlit run app.py

---

🛠️ Usage Guide

1. Enter or Edit Paragraphs

• In the left column, type or paste your paragraph.
• Click "Generate Graph" to visualize the concept graph for that paragraph.
• A node wise Reading Direction will be generated to ease reading.
• Node numbers are displayed in he start of node labels in the graph

2. Adjust Physics Controls

• Use the sliders in the rightmost column to adjust the appearance and layout of the graph:
  • Spring Length: Controls the gap between nodes.
  • Spring Constant: Adjusts flexibility.
  • Gravitational Constant: Controls clustering.
  • Central Gravity: Pulls nodes toward the center.

3. Add More Paragraphs

• Click "Add Paragraph" to insert another text box and repeat the process.

4. Combine Information

• Click "Article Covered: Combine Information" upon entering all paragraph to generate a combined concept graph of all key sentences of the passage.
• Use the left column to adjust physics controls for the combined graph.
• An Article Walkthough in directed graph fashion will be displayed
• Download the combined graph as HTML.

5. Explore and Download

• Interact with the graphs directly in the app.
• Download any graph for offline viewing or sharing.

---

💡 Example Use Cases

• Visualizing logical flow in essays, articles, or research papers
• Exploring conceptual connections in academic writing
• Teaching or learning about graph-based text analysis
• Optimizing the structure of technical or scientific documents

---

🧑‍💻 Development & Contribution

Contributing

Contributions are welcome!

• Fork this repo and submit a pull request.
• Please open an issue for bugs, suggestions, or feature requests.

Coding Style

• Follows PEP8 and Streamlit best practices.
• Modular, readable code for easy extension.

Testing

• Test locally with a variety of text inputs and paragraph counts.
• Check graph interactivity and download features.

---

📣 Feedback & Support

• Found a bug or want to suggest a feature?
  Open an issue or submit a pull request!
• For questions or functioning related queries, mail me: Aadityaamlan Panda

---

🤝 Credits

• Built with Streamlit
• NLP by NLTK, scikit-learn
• Graphs by NetworkX and PyVis

---

📄 License

This project is licensed under the MIT License. See LICENSE for details.

---

🔗 References

• Streamlit Docs: Creating and Sharing Apps
• Streamlit Deployment Guide
• Awesome Streamlit Project Templates
• GeeksforGeeks: NLTK Tutorial

---

Enjoy exploring your ideas visually!