OntoBot
OntoBot Documentation
OntoBot is a production-ready, end-to-end platform for human–building conversation in natural language. It combines Rasa 3.6.12, analytics microservices, knowledge stores (SQL/SPARQL/NoSQL), and optional AI helpers to turn sensor and ontology data into user-friendly insights.
🏢 Building-Specific Documentation
Explore documentation for each of our three smart buildings:
- Building 1 - ABACWS - Real university testbed with 680 sensors, MySQL, Indoor Environmental Quality (IEQ) monitoring
- Building 2 - Office Building - Synthetic commercial building with 329 sensors, TimescaleDB, HVAC optimization
- Building 3 - Data Center - Critical infrastructure with 597 sensors, Cassandra, cooling & power monitoring
Total: 1,606 sensors across all buildings
📚 Getting Started
Quick Start & Installation
- Quick Start Guide - Get OntoBot running in 30 minutes
- Installation Guide - Detailed setup instructions
- Architecture Overview - System design and components
Core Documentation
- Introduction - What is OntoBot and key features
- Usage Guide - How to interact with the system
- Services Overview - All 8+ microservices explained
🔧 Technical Documentation
Database Integration
- Database Integration Guide - MySQL, TimescaleDB, and Cassandra implementation guides
- MySQL 8.0 (Building 1) - Traditional relational database
- TimescaleDB 2.11 (Building 2) - Time-series optimized PostgreSQL
- Cassandra 4.1 (Building 3) - Distributed NoSQL for high availability
Advanced Topics
- Backend Services - All 8 services with API references
- Frontend UI Guide - Complete UI reference and keyboard shortcuts
- T5 Training Guide - NL2SPARQL model training workflow
- Multi-Building Support - Switching buildings and portability
- Analytics API Reference - 30+ analytics types with examples
- Data & Payloads - Request/response formats
- Customization Guide - Extending OntoBot for your needs
🐛 Support & Troubleshooting
- Troubleshooting & FAQ - Common issues and solutions
- Docker & Compose issues
- Database problems (MySQL, TimescaleDB, Cassandra)
- Service health issues
- Performance optimization
- Development workflow
- Testing & Operations - Testing procedures and operational best practices
📊 Features by Building
| Feature | Building 1 | Building 2 | Building 3 |
|---|---|---|---|
| Sensors | 680 (IEQ) | 329 (HVAC) | 597 (Critical) |
| Database | MySQL 8.0 | TimescaleDB 2.11 | Cassandra 4.1 |
| Focus | Air Quality | Thermal Comfort | Cooling & Power |
| Type | Real Testbed | Synthetic Office | Synthetic DC |
| Special Features | CO2, PM, TVOC | AHUs, Chillers, VAV | CRAC, UPS, PDU |
🚀 Technology Stack
- Conversational AI: Rasa 3.6.12
- Backend: Python 3.10, Flask, FastAPI
- Frontend: React 18+
- Databases: MySQL 8.0, TimescaleDB 2.11, Cassandra 4.1
- Knowledge Graph: Apache Jena Fuseki, Brick Schema 1.3
- Analytics: 30+ analytics types (forecasting, anomaly detection, etc.)
- NL2SPARQL: T5-based transformer model
- LLM: Mistral via Ollama for summarization
- Containerization: Docker & Docker Compose
📖 Documentation Index
By Category
Buildings & Infrastructure:
- Building 1 - ABACWS (Real Testbed)
- Building 2 - Office (HVAC Focus)
- Building 3 - Data Center (Critical Infrastructure)
- Multi-Building Support
Database & Integration:
Services & APIs:
User Interfaces:
Development & Operations:
- Quick Start
- Installation
- Architecture
- Usage
- Customization
- Testing & Operations
- Troubleshooting & FAQ
🎓 Research & Academic Use
OntoBot is designed for both production deployment and academic research:
- Real-world testbed data (Building 1 - ABACWS at Cardiff University)
- Synthetic building scenarios (Buildings 2 & 3 for controlled experiments)
- Multiple database technologies for comparative studies
- Open-source and extensively documented
- Reproducible experiments with Docker Compose
🤝 Contributing
OntoBot is actively developed and welcomes contributions:
- GitHub Repository: suhasdevmane/OntoBot
- Issues & Bug Reports: GitHub Issues
- Documentation: This site!
📄 License
OntoBot is released under the MIT License. See the LICENSE file for details.
🔗 External Resources
- Rasa Documentation: rasa.com/docs
- Brick Schema: brickschema.org
- TimescaleDB: docs.timescale.com
- Cassandra: cassandra.apache.org
- Docker: docs.docker.com
Ready to get started? Check out the Quick Start Guide to get OntoBot running in 30 minutes!