UD-GraphClustering

These demonstrations are part of a student research project for visualizing and navigating graphical data on the web. This project will use a web-based data visualization framework to visualize knowledge graphs.

Installation

The following sections detail how to setup the demo. In case of crash or error where the demo containers are down, it is possible that the data may need to be reuploaded to the blazegraph container.

Pre-requisites

• Install Docker

Repository Setup

Currently, the UD-Viz framework must be initialized after cloning this repository.

git clone https://github.com/VCityTeam/UD-GraphClustering.git
cd UD-GraphClustering

To install the demo:

npm install
npm run build

Component Setup

To configure the demo and the components that support it, edit the .env file to be launched with docker-compose. By default, the following port are used by the following services:

• 9011: Blazegraph
• 9012 : UD-Viz - demo sparql widget
• 9013 : UD-Viz - demo json data

Build Images, Demos and Run Containers

First, make sure to set the sparqlModule/url port in the sparql_widget.json file here to the same value as the BLAZEGRAPH_PORT variable declared in the .env file.

Then build the Blazegraph docker image, build both demos and run their containers:

docker compose up

Upload RDF-Store Dataset

To upload files into the RDF-store to be used by the sparqlModule, you can use the online Blazegraph interface here.

Make sure that you've uploaded the corresponding 3DTiles layer in the 3DTiles_temporal.json file here in order to visualize the 3D scene.

Now the demo UD-Viz - demo sparql widget and UD-Viz - demo json data are ready and can be accessed from localhost:9012 and localhost:9013.