DiMad Knowledge Graph Access with Python

This repository contains a Jupyter notebook Workflow.ipynb that demonstrates how to access and explore measurement data from the DiMad Knowledge Graph using Python. DiMad is a research project investigating wire-based additive manufacturing of high-alloy steels. For more information, visit the MaterialDigital website.

The DiMad ontology, developed as part of this effort, is published as an OWL (Web Ontology Language) graph and hosted on a public SPARQL endpoint. While SPARQL is a powerful query language for accessing semantic data, it can be non-trivial to use - especially for domain experts without prior experience in semantic technologies.

To mitigate this barrier, our notebook uses SPARQL only for the initial data retrieval, after which the pandas library is used for data manipulation and visualization. This approach is practical given the manageable size of the DiMad Knowledge Graph, which can be loaded entirely into memory for exploratory analysis.

Requirements

The notebook requires the following Python packages:

• SPARQLWrapper for querying the DiMad Knowledge Graph
• pandas for data manipulation
• matplotlib for data visualization
• ipywidgets for interactive widgets

You can install these packages using pip:

pip install SPARQLWrapper pandas matplotlib ipywidgets

Usage

To run the notebook, open it in Jupyter and execute the cells in order. The notebook will guide you through the process of querying the DiMad Knowledge Graph, loading the data into a pandas DataFrame, and exploring the measurement results.