Mining cells of natural language (for semantic ontology)

This is a brief but relevant comment, from my Systems Thinking World discussions, points out a way the efforts by Goggle and others, to mine “meaning” from of the massive quantities of semantic data now available, is missing a golden opportunity. There are a variety of ways to use the natural structures of languages as a key.

@Ferenc – I don’t recall the subject of data mining semantic meaning coming up, but I sure agree there seems no computer search strategy yet in use for that.    I have some original technical ideas of how to do it, but they all begin with  learning to recognize how natural languages “integrate common knowledge” for you, by how language communities naturally develop within their own social commons, (or “silo”).

So the first step to learning how to read the natural organization of semantic structures generally is to learn how recognize and observe the development of natural languages and the semantic webs they create.   This STW community is one, for example, as is any other community with a sustained internal conversation.

Armed with that, perhaps a computer whiz could learn to crawl the web to develop a lexicon of the code phrases of a great variety of distinctive language communities.  That could provide a way to let you search on any topic of your interest, for any language group’s interest in it.

I’ve tried to suggest that to Google a few times, to let people do web searches from a “scientific” viewpoint, or “entertainment” or “youth” or “religious”, “liberal”, “conservative”, “European”, “Asian” or other distinctive community of interest.

Wouldn’t having that option, to look in on other language cultures and learn from what they’re learning from, would be very entertaining and enlightening itself, wouldn’t it?

—-

also

The physics of HappeningStatistical Methods

 

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.