/temi-kg-ui

Knowledge graph UI code used in our CHI'21 paper "Patterns for Representing Knowledge Graphs to Communicate Situational Knowledge of Service Robots".

Primary LanguageJavaScript

Knowledge Graph UI for the Temi robot

Knowledge graph UI code used in our CHI'21 paper "Patterns for Representing Knowledge Graphs to Communicate Situational Knowledge of Service Robots".

A screenshot of the KG UI.

Running the code

Set up the knowledge graph

After you've set up an empty database as described in this article, download the neo4j graph data dump we have prepared here. You can import the data by putting it in you database folder and running the neo4j-admin command:

neo4j-admin load --from=lab-kg.dump --database=graph.db --force

More detailed instruction can be found in the Neo4j Docs

Install python libraries

python -m pip install flask neo4j

Start python server

python server.py

Code Structure

  • server.py is the main entry point of the program. It starts a Flask API for KG queries, and serves the UI code in the static folder.
  • kg.py contains a class that handles the communication between Neo4j and Python. The Flask server uses this class to talk to Neo4j.
  • static contains code for a web-based GUI built with Vue.