TextScape

A Confabulation-Inspired Text Knowledge Explorer

TextScape transforms your written content into a navigable knowledge graph, leveraging principles from confabulation theory to uncover hidden connections and patterns.

---

Features

• Semantic Chunking: Automatically breaks text into discrete "knowledge units" (paragraphs or sections).
• Confabulation-Inspired Associations: Builds weighted links based on co-occurrence strength and semantic similarity.
• Activation Spreading: Select a node to simulate neural activation and reveal related concepts.
• Interactive Visualization: Explore your corpus via a force-directed graph (D3.js) with zoom, pan, and drag.
• Search & Highlight: Find nodes matching queries and highlight them in the graph.

---

Getting Started

Prerequisites

• Modern web browser (Chrome, Firefox, Brave, Edge)
• Git & GitHub account

Installation & Setup

1. Clone this repo

   git clone https://github.com/Innomen/textscape.git
   cd textscape

2. Open in browser

   • Double-click text_scape.html or serve locally:

     # with Python 3
     python -m http.server 8000
     open http://localhost:8000/text_scape.html

Usage

• Upload .txt or .md files via the sidebar.
• Add text directly in the text area.
• Click a node to view full content and related concepts.
• Search to highlight matching nodes.
• Drag/Zoom to navigate large graphs.

---

Architecture

• Frontend: HTML + D3.js for visualization, plain JavaScript modules
• Text Processing: Simple tokenizer & paragraph splitter
• Association Engine: Basic TF-style embeddings & cosine similarity
• Activation Model: Spreading activation based on link strength

For full details, see comments in text_scape.html.

---

Roadmap

• Integrate advanced NLP (spaCy, embeddings)
• Support additional formats (HTML, PDF)
• Server-side processing for large corpora
• Persistent storage (MongoDB, Graph DB)

---

Contributing

1. Fork the repo
2. Create a feature branch
3. Submit a pull request

---

License

MIT License © Innomen