This website presents a proof-of-concept search engine system for accessing chemical data
from the World Avatar Knowledge Graph and the Wikidata Knowledge Graph. The Knowledge Graphs
offers inter-connected data from chemical kinetics to chemical and physical properties of
species and many other domains. We modelled the chemistry knowledge using various knowledge
graph embedding techniques and leverage pre-trained BERT model to interpret questions posed
by users. This website is a demonstration of the system described by the preprint "Marie and BERT".
Marie is a question answering system for knowledge graphs with a natural language interface. If the user would like
to query OntoSpecies using SPARQL, please follow this link.
Please note that Marie is still under development and the accuracy of the results will increase
with further refinement of the underlying ontologies.
To use the Marie search engine, type a question into the field below, or select one of the provided sample questions.
CMCL Innovations cmclinnovations.com |
Computational Modelling Group como.ceb.cam.ac.uk |
Cambridge CARES cares.cam.ac.uk |