Free and open thesaurus to include in the package

[njtierney]

Possible sources:

gutenberg.org/ebooks/10681 (Thanks Chester Ismay!)

[Lingtax]

This SO thread might have a few options: https://stackoverflow.com/questions/5618304/looking-for-thesaurus-data

Some are now dead, but wordnet might help (https://wordnet.princeton.edu/); and if you could parse the structure, the open office dictionaries also.

[njtierney]

Thanks @Lingtax ! Appreciate it :)

Here are some other notes for me to look at

http://thesaurus.altervista.org/
Google search
https://old.datahub.io/dataset/open-data-thesaurus
https://www.quora.com/Thesaurus-Ontologies-is-there-a-downloadable-database-of-the-english-thesaurus-for-my-own-use
https://en.wikipedia.org/wiki/OpenThesaurus

[njtierney]

I could also consider wrapping a JS library, like moby - https://github.com/words/moby#readme

[njtierney]

theres the mythes from the hunspell repo:

https://github.com/hunspell/mythes

This one looks like a winner!

[njtierney]

We have currently used https://github.com/words/moby

[njtierney]

@coolbutuseless I wonder if the hunspell/mythes would be a better alternative in the future?

[coolbutuseless]

That might depend on how keen you are to link in a C++ library to solve this problem!

I think it might be overkill, but some of the stemming stuff seems pretty useful.