05: Conversation History¶
Overview¶
Shows how AI maintains and evolves its understanding of a user through multi-turn conversation. Each turn adds or refines knowledge, demonstrating incremental memory building in dialogue systems.
Theme¶
Conversational Memory Evolution
Strategy¶
MERGE_FIELD with incremental fact accumulation
Key Features¶
- ✅ Turn-by-turn memory updates
- ✅ Incremental fact accumulation
- ✅ Automatic context maintenance
- ✅ Memory-aware response generation
- ✅ User preference tracking
Data Structure¶
ConversationMemory¶
session_id: str # Conversation session ID
user_name: str | None # User's name
known_topics: list[str] # Discussed topics
user_preferences: list[str] # Stated preferences
user_interests: list[str] # Inferred interests
communication_style: str | None # Preferred style
problem_history: list[str] # Previous issues
solutions_applied: list[str] # What's been tried
session_count: int # Number of sessions
last_interaction: str | None # Last update time
Use Case¶
Conversational AI & Chatbots: Build chatbots and dialogue systems that remember user preferences, track conversation history, and improve responses over time through accumulated knowledge.
Benefits: - Personalized responses based on history - Reduced need to re-explain context - Progressive trust building - Improved user experience
Running the Example¶
cd examples/
python 05_conversation_history.py
Output¶
Results are stored in temp/conversation_memory/:
- memory.json: Conversation memory records
- metadata.json: Schema and statistics
No API Required ✅¶
This example works without any external API keys or dependencies beyond the core OntoMem package.
What You'll Learn¶
- Multi-Turn Updates: Updating memory across conversation turns
- Incremental Accumulation: Building profiles gradually through dialogue
- Context Maintenance: Keeping track of conversation context
- Preference Tracking: Recording user preferences and interests
- Dialogue Integration: Integrating OntoMem with chat systems
Complexity¶
⭐⭐⭐ Intermediate: Shows practical dialogue system integration patterns.
Related Concepts¶
Next Examples¶
- 02: RPG NPC Memory - Similar profile building in gaming
- 06: Temporal Memory - Time-based aggregation