:: AForge.NET Framework :: Articles :: Forums ::

Working with Text

The forum is to discuss topics from different artificial intelligence areas, like neural networks, genetic algorithms, machine learning, etc.

Working with Text

Postby karell » Tue Apr 06, 2010 3:58 pm

Hello Folks,

I continue to perform experiments with the library to further my goals and research, I am now at the point of asking this question:

What features of the library can I use in order to work with non-quantified values? (Not numbers)

Say text...

I have a paragraph of text that I want to be able to analyze in order to understand some basics around it (meaning/purpose/etc) and then be able to understand if this information is relevant to others and then perhaps be able to rate the information according to desirability of the user to use/see it.

As examples:
    Does the director of IT want to see the resume of a person that has a background in sales?
    Where should a Request for Proposal go to when it comes in through a general-purpose e-mail address?
    When I consult craigslist, what adds do I want to see?

This is a very interesting "advanced" problem and while I'm quite certain that the library doesn't yet address Corpus Linguistics, I would like to get input from anyone that has any leads of doing similar things, especially in the .NET world.

Any leads would be appreciated so that I can find a direction.

Karell Ste-Marie
CIO BrainBank Inc
User avatar
Posts: 6
Joined: Mon Feb 22, 2010 5:50 pm
Location: Montreal, Quebec, Canada

Re: Working with Text

Postby andrew.kirillov » Tue Apr 06, 2010 4:30 pm

Hello, Karell

I am not really sure there will be much useful in the AForge.NET framework for such task like analyzing text and understanding its meaning. So you will definitely need to continue your search for something else.

I do remember that Creating Artificial Intelligence Google group had quite a lot discussion related to text analysis. So you may try it there.
User avatar
Site Admin, AForge.NET Developer
Posts: 3453
Joined: Fri Jan 23, 2009 9:12 am
Location: UK

Return to Artificial Intelligence