Skip to content

User Context Data Model

Personalization is only as good as the data model behind it. This interactive graph shows how a chatbot organizes everything it knows about a user: a central User node linked to a profile, preferences, history, and active sessions, which in turn branch into individual queries, settings, and learned behavioral patterns.

Interactive Demo

Run MicroSim Fullscreen

You can embed this MicroSim in your own page with the following iframe:

1
<iframe src="main.html" width="100%" height="642" scrolling="no"></iframe>

Overview

The model centers on the example user Alice Chen and uses color to group node types: User (purple), Profile (pink), Preferences (blue), History (green), Sessions (orange), Queries (gray), Settings (light blue), and Patterns (light green). Edges are labeled with the relationship type and styled by strength:

  • Thick solid — HAS_PROFILE, HAS_PREFERENCES, HAS_HISTORY
  • Medium solid — INITIATED (User to Session)
  • Thin solid — CONTAINS (Session to Query)
  • Dashed — CONFIGURED_BY and EXHIBITS (derived attributes)

Interactions:

  • Hover or click any node to see its full property list in the side panel.
  • Click a node to highlight its connected edges (others dim).
  • Use the checkboxes to hide or show sessions/queries, settings, or behavioral patterns and focus on one part of the model.
  • Use the navigation buttons to pan and zoom; in fullscreen mode the mouse wheel zooms too.

Lesson Plan

  • Read the graph: Have students name the path from the User node to a single query and list every relationship label along the way.
  • Explore: Hover each node type and record one property that would be useful for personalization.
  • Analyze: Discuss why CONFIGURED_BY and EXHIBITS are dashed (derived) while HAS_PROFILE is solid (direct).
  • Design: Ask students to add a new node type, such as a "Device" node, and decide how it connects to the User.

References