Materials science experiments involve complex data that is hard to reproduce and might sometimes be redundant. It was observed in specific cases where the data from materials sciences engineering (MSE) experiments were generated using linked open data principles[1,2], e.g., the RDF data model. To query this data, detailed knowledge of formulating SPARQL queries is necessary.

The problem observed was a need for more knowledge in SPARQL to query this data by domain experts.

With this work, we aim to develop an approach to the domain expert where instead of SPARQL queries, the user can develop expressions in Natural Language, e.g., English, to query the data. We plan to evaluate our approach, with a different number of data, from different sources. We want to compare with synthetic data to assess the quality of the translations of our approach. We also aim to benchmark the state-of-the-art approaches to measure the efficiency of the methods, as mentioned earlier, with specific indicators to find their barriers to introducing a new approach to cover the problems and provide a better solution.

References

[1] T Berners-Lee, J Hendler, O Lassila - Scientific American, 2001, 284, 34–43.

[2] RDF specification. 2023. available at: https://www.w3.org/RDF/

[3] Bundesanstalt für Materialforschung und -prüfung (BAM) 2022. Available at: https://www.bam.de/

[4] Mat-o-Lab. 2023. Available at: https://github.com/Mat-O-Lab.