Apertium is a Free Software system for rule-based machine translation. It is primarily aimed at pairs of languages. For every language pair it includes a monolingual dictionary for each language, a bilingual dictionary that maps word correspondences between languages, and lists of grammatical rules for transforming one language into another. It stores such information in files of its own format. Modifying these files at the moment requires skills with programming, source code control systems, compilation, etc.
It sounds like Lexical Wikidata stores a lot of information that can be used almost directly in Apertium, and editing this information will be as easy as editing Wikidata. This is probably easier that editing source code files, so it would be possible to use Lexical Wikidata to empower the crowdsourcing of building dictionaries for Apertium.
(Apertium is one particular open source machine translation system that I know, but it can also be used by any other comparable system.)