metaphactory could be improved in terms of data integration. For now, you can design ontologies and taxonomies, but you cannot directly implement, natively, the integration of data associated with the different taxonomies, which makes the integration of mapping engines like R2RML very useful. It would be interesting if metaphactory could also integrate engines for creating inference rules, because the goal of knowledge graphs is to create inference rules, and managing inference rules as well would be interesting. Regarding metaphactory’s artificial intelligence capabilities, I would say that governance and data security is something that for me is not integrated. Overall, when I look at metaphactory, I don’t think of security; I think of integration, and while there is access management, security is something else. At the beginning, the AI features were really just tests, and the results, in terms of reliability and accuracy, were not really there yet, remaining quite tentative. Since then, I haven’t had the chance to test the AI side further. I do not have any other improvements to suggest.
