What is needed in order to “get” information by keywords?

Question

Machine learning

Hey!

I'm just starting to learn ML + AI - like many others, I'm at the very beginning of the path - in connection with which I had a problem.

I would like to at least roughly understand or find out what can be read on the topic - if I need to do something like this:

by keywords, so that everything that at least somehow fits in the meaning, for example:
the keyword is "PURITY" --- > what I get in response: SOAP, POWDER, WATER, MOP, HYGIENE and so on in the same vein. It is not at all necessary that we are talking about individual words - at least just some kind of "web scraping" for concepts related to the keyword.

Maybe - this algorithm already has a name?

In general, I would be very grateful for ANY help in this matter. At least understand where to dance.

Thanks a lot.

Answer 1

This question has nothing to do with ML. And it refers to linguistics, where there is such a thing as "ontology" and the so-called. " Ontological
Graphs " _
_
.ru/studies/courses/1078/270/info -- more fundamental.Dig
in this direction.Only there is no universal solution to such a problem, and as a rule, for each subject area and even for each small part of the subject area, they build their own ontological scheme.Even in your example, "purity" can include both "purity of the experiment" and "purity of the compound", and "purity of the diamond", etc., etc.