How Can Associative Semantic Nets Revolutionize Business and Research?

  • Thread starter Thread starter fxseminar
  • Start date Start date
  • Tags Tags
    associative
AI Thread Summary
The discussion centers on Associative Semantic Nets, distinguishing them from neural networks by emphasizing that neural nets are algorithms while semantic nets serve as a presentation of semantic relationships between concepts in natural language. One participant is developing a tool that creates Associative Semantic Nets through statistical content analysis, with an example provided at SemantiCat.com. The tool aims to produce specialized nets for various topics. The conversation touches on the structure of semantic nets, where nodes represent words and ribs indicate associative relationships between them. Participants discuss the possibility of assigning values to these associations and the varying types of relationships that can exist within the nets. The developer expresses interest in the project for personal, academic, and business reasons, highlighting the potential of Associative Semantic Nets in various applications.
fxseminar
Is anybody here interested in Associative Semantic Nets?


.
 
Physics news on Phys.org
neural nets? sure why not.
 
No, it is not neural net.

Neural net is ALGORITHM while Semantic Nets is PRESENTATION.

Presentation of semantic relations between "concepts" (=words) in the natural language.
 
Okay, I'm interested. Now what? :cool:
 
Now - fine.

Are you developing some project conserned?
 
I'm intregued by general semantics which I have read a bit on. Not the same thing really I'm sure. So are you working on a project concerning this?
 
Yes, I do. I develop a tool producing Associative Semantic Nets by statistical analysis of content.

The example you can see at SemantiCat.com where the Net is used to be a structure of a site.

While that Net is "universal" I'm preparing to produce specialized ones for any given topic.
 
I'm not yet working on any project with them; I just looked up a basic definition. I am studying linguistics but haven't gotten to semantics yet. Semantic nets look like they use logic that I'm familiar with.
 
Logic of semantic nets is based on logic of ANY net (graph): base unit is - two nodes and rib between them.

Then we must interpret what is nodes and what is ribs.

As semantic for us is the branch of linguistics so nodes here are words.

Particularly in ASSOCIATIVE semantic nets rib serves as fixations of the fact of associative relation between words the rib connects.
 
  • #10
Do the ribs themselves represent words? Or are there are few fixed types of relations? I saw x-->instance-->y or Instance(x, y) used a lot. Just glanced through it though.
 
  • #11
They can - in some kinds of semantic nets.
You need to assign some "text" to the rib - only if there are DIFFERENT kinds of ribs in your net.

If ALL ribs means "instance of" then you can simply title your net "instance semantic net".

The same way - in ASSOCIATIVE semantic net sense of all the ribs is "association" between words.

By the way there is another feature: VALUE of association. So in Associative Semantic Net rib is assigned with number, nor "text"!
 
  • #12
Okay, cool. So are you doing this for school, business, fun, ...?
 
  • #13
I'm doing this for a set of reasons:

1. it is interesting for me
2. I guess it's perspective
3. I want to make business on it.
 
Back
Top