Glossary

Glossary

This glossary explains the key terms and concepts we use throughout this site. Understanding these concepts will help you get the most out of our interactive demos and methodology.

AI Assistant

An AI Assistant is an intelligent software agent that can understand natural language questions and provide relevant answers based on a knowledge base. In our case, the AI Assistant uses a knowledge graph as its knowledge base and can answer questions about enterprise structures, processes, and relationships.

Key capabilities:

  • Understands questions in natural language
  • Queries knowledge graphs to find relevant information
  • Provides contextual answers with supporting details
  • Can explore complex relationships across multiple entities

Example: “Which systems support the customer service department?” - The AI Assistant queries the knowledge graph and returns all systems connected to customer service.

Generative AI

Generative AI refers to artificial intelligence systems that can create new content—text, images, code, or other outputs—based on patterns learned from training data. Large Language Models (LLMs) like GPT-4 or Claude are examples of generative AI.

Key characteristics:

  • Creates human-like responses to questions
  • Understands context and nuance in conversations
  • Can explain complex concepts in simple terms
  • Works best when combined with accurate, domain-specific knowledge

In our context: We use generative AI to power conversational interfaces that make enterprise knowledge accessible through natural dialogue.

Knowledge Graph

A Knowledge Graph is a structured representation of information that models entities (things, concepts, or objects) and the relationships between them. Unlike traditional databases that store data in tables, knowledge graphs store data as nodes (entities) and edges (relationships), making it easy to traverse connections and discover patterns.

Key features:

  • Entities: Individual objects or concepts (e.g., “Customer Service Department”, “CRM System”)
  • Relationships: Connections between entities (e.g., “uses”, “supports”, “manages”)
  • Properties: Attributes of entities (e.g., name, description, status)
  • Schema flexibility: Easy to extend with new entity types and relationships

Benefits:

  • Reveals hidden connections across your organization
  • Supports complex queries across multiple dimensions
  • Enables graph algorithms for analysis and optimization
  • Provides context for AI-powered insights

Example: In our car sharing demo, the knowledge graph connects departments → processes → systems → goals, allowing you to ask “How does the fleet control system support our CO₂ reduction goals?”

Ontology

An Ontology is a formal specification of the concepts, entities, and relationships within a specific domain. It defines what things exist, what properties they have, and how they relate to each other. Think of it as a blueprint or schema for organizing knowledge.

Key components:

  • Classes: Categories of things (e.g., “Process”, “System”, “Goal”)
  • Properties: Attributes that things can have (e.g., “name”, “owner”, “status”)
  • Relationships: How things connect (e.g., “implements”, “supports”, “delivers”)
  • Rules: Constraints and logic about how things work together

Difference from Knowledge Graph:

  • Ontology = The schema/structure (what can exist and how)
  • Knowledge Graph = The actual data (what does exist)

Example: Our car sharing ontology defines that “Processes are implemented by Systems” and “Goals are achieved by Processes”. The knowledge graph then contains specific instances like “Car Rental Process is implemented by Renting System”.

Retrieval Augmented Generation (RAG)

Retrieval Augmented Generation (RAG) is a technique that combines the power of generative AI with the accuracy of information retrieval. Instead of relying solely on the AI’s training data, RAG first retrieves relevant information from a knowledge base (like a knowledge graph), then uses that information to generate accurate, contextual responses.

How it works:

  1. User asks a question → “What processes does the Customer Service department handle?”
  2. Retrieval step → Query the knowledge graph to find relevant entities and relationships
  3. Augmentation step → Provide the retrieved information as context to the AI
  4. Generation step → The AI generates a natural language answer based on the retrieved facts

Benefits:

  • Accuracy: Answers are grounded in actual enterprise data, not AI hallucinations
  • Current: Knowledge is always up-to-date with your latest data
  • Traceable: Answers can be linked back to source entities in the knowledge graph
  • Domain-specific: AI has deep knowledge of your specific enterprise

Our implementation: When you chat with our AI Assistant, it uses RAG to query the Neo4j knowledge graph, retrieve relevant enterprise elements, and generate accurate answers based on your actual business model.

How These Concepts Work Together

In our AI-powered enterprise transformation approach:

  1. Ontology provides the structure for your enterprise knowledge
  2. Knowledge Graph stores your actual enterprise data according to that structure
  3. RAG retrieves relevant information from the knowledge graph when needed
  4. Generative AI creates natural, conversational responses
  5. AI Assistant brings it all together in an interactive experience

This combination enables you to have natural conversations with your enterprise data, uncovering insights and connections that would be difficult to discover with traditional tools.

Further Reading

How it Works