Contextual-Memory-Graph-System
.png)
🧠 Advanced AI Knowledge Graph with Contextual Memory
Mobile-optimized system that builds dynamic relationships from conversations and documents
🚀 What It Does
- 🔗 Builds Knowledge Graphs - Automatically extracts entities and relationships from text
- 🧠 Contextual Memory - Preserves meaning across interactions, not just keywords
- 📱 Mobile-First - Optimized for Android devices with efficient SQLite storage
- 🔒 Privacy-Preserving - All processing happens locally on your device
⚡ Quick Start
# Download and run
python contextual_memory_mvp.py
# Add memories
cms.add_memory("John works at Google and specializes in AI research.")
# Query the system
results = cms.query_memory("Tell me about AI research")
🎯 Key Features
| Feature | Description |
|---|---|
| Entity Extraction | Automatically identifies people, organizations, concepts |
| Relationship Detection | Discovers connections like “works at”, “created”, “uses” |
| Semantic Search | Context-aware retrieval beyond keyword matching |
| Importance Scoring | Memory relevance evolves based on access patterns |
🏗️ Architecture
ContextualMemorySystem
├── NLP Processor (Entity extraction, relationships)
├── Context Weaver (Memory creation, linking)
├── Memory Retriever (Query processing, ranking)
└── Mobile Storage (SQLite backend, graph relations)
📊 Demo Results
System Stats: {'entities': 48, 'relationships': 15, 'memories': 5}
Query: 'Tell me about John'
1. [Score: 0.348] John works at Google and specializes in AI research...
2. [Score: 0.348] The AI algorithm that John developed uses Python...
✓ Database created: demo_knowledge.db (77,824 bytes)
🎯 Use Cases
- Personal AI Assistant - Build personalized knowledge from conversations
- Enterprise Knowledge Management - Extract insights from company documents
- Research & Development - Track connections and dependencies
- Education & Learning - Create personalized learning paths
🛡️ Privacy & Security
- 🔒 Local Processing - All data stays on device
- 🚫 No External APIs - No data transmitted to external services
- 🔐 Encrypted Storage - SQLite database with optional encryption
📱 Requirements
- Python 3.7+
- Standard library only (no external dependencies)
- Tested on Samsung Galaxy S24 with Pydroid 3
📁 Files
contextual_memory_mvp.py- Main system implementationREADME.md- Complete technical documentationLICENSE- MIT License
👤 About the Author
Justin Lane
🔗 GitHub: @aiwithjusl
🔗 LinkedIn: Justin Lane
📬 Email: aiwithjusl.dev@gmail.com
**⭐ Star this repo if you find it useful! ⭐**
*Built for senior-level AI/ML engineering positions and enterprise consulting opportunities.*