• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans


Adverts help to support the work of this non-profit organisation. To go ad-free join as a Member.


Photo
- - - - -

Data mining forums


  • Please log in to reply
No replies to this topic

#1 caston

  • Guest
  • 2,141 posts
  • 23
  • Location:Perth Australia

Posted 15 April 2010 - 03:05 PM


Perhaps one of the reasons why the development of the imminst wiki has stalled is that its such a tremendous task that it scares people away.

I was doing a search on forums vs blogs vs wikis on google the other day and I found some interesting insights


"Fundamentally, forums and wikis serve different purposes. Wikis are good for creating a relatively compact knowledgebase, but aren't great for building a community. Forums are much better for allowing members to interact with each other, but the knowledge tends to be dispersed and somewhat lower in density."

http://www.webmaster...orum103/898.htm post: #:1562468

or

"Forum vs Blog vs Wiki
I've pondered a few times the benefits of various tools as a means of building an internet community.

Forums: Get lots of momentum and conversation happening, but things tend to vanish into the archives and get lost. Topics can also get off-track quickly
Blogs: More directed than forums, but less flexible.
Wikis: Most flexible, great ability to cross-link information. Potential to be ruined by vandalism / revert wars. Less structured format doesn't lend itself to debate / discussion / conversation as well, but can result in a more coherent final position.
I feel that, overall, wikis have the potential to generate the best results. Perhaps recent changes can be highlighted and grouped in some way to appear more like forums. Maybe some user-education would help to make people feel more comfortable."

http://www.yourdemoc...drupal/node/870


There are so many useful nuggets of information out there on forums but they are constantly being buried under new posts and threads.
If you are looking for one simple thing e.g. you want to find all the posts on facial exercises you can search for "facial exercises" but if you want to build a knowledge base based on every conceivable topic and build it into to a wiki it would take a lot of man hours either yours alone or many many people each contributing if you can start such a movement.

I was wondering if it fact it may be possible to develop some basic machine learning data mining based algorithms that could actually do the grunt work in sorting most of the forum posts into the correct topics based on keywords and association. From there it would be easier for a human to further refine the information and start building a wiki article for each topic.

There are some open source data mining applications e.g. Rapid Miner but I don't know if there are any that would be specifically good for data mining forums. Perhaps this is more of a question for the singinst crowd but I would like to raise it here first as this is the community I feel most at home in.

If we come up with a good solution we could datamine other forums as well and create an ecosystem of wikis built using high quality information nuggets found in forums and that share content with each other.

Edited by caston, 15 April 2010 - 03:07 PM.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users