is there a model or method using Knime where a ML solution could be presented
with a document and would (based on English-English (as oppose to US english)
highlight and/or fix spelling, highlight and/or fix complex sentences.
I do a lot of work with IT engineers who are brilliant at their job, but not
great at the written language. I’d love to build a ML model that could learn
based on the documents they are producing in a given body of work such as a
project where the same terms etc are used over and over to fix the written
work and to also pick up where in one document we may have said one thing
(e.g. how a given part of the solution will be built) and then contradicted
that in some other document.
This would be a great way to remove defects from documents and from the build
process.
Hi @dreisner -
This sounds like an interesting project, but also one that is fairly complex. I don’t know offhand of any out-of-the-box workflows we have available on the Hub that would do this. Certainly searching for particular terms and tagging or making replacements could be done, but free text generation and editing would be highly specialized and lots of work…
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.