Techmeme/Google News style grouping of related items?

[milk]

Hey all,

I'm new to things and wondering if SwiftRiver has any capabilities for
automatic clustering of news articles (beyond basic auto-tagging,
semantic or not), along the lines of Techmeme - http://techmeme.com
(or it's later politics related sibling site Memeorandum -
http://www.memeorandum.com/) or Google News? (n.b. the Techmeme system
is part algorithm, part curated)

The closest open source tool I've found to this so far is a GSOC
project to create a Drupal Memetracker module -
http://kyle.mathews2000.com/blog/2008/04/04/drupal-memetracker-module-my-google-summer-of-code-application
- http://drupal.org/project/memetracker - I've yet to dive into that,
but that's looking very quiet these days.. Any recommendations for a
similar system out there? I feel a tool like this would great help for
the day-to-day tracking of specific elements in the sometimes fast
moving narratives relating to an event or topic so users can then
apply their own knowledge and skills to curating the 'display flow' of
said items. Thanks for any advice!

-milk

[M. Edward (Ed) Borasky]

On Fri, 7 Jan 2011 05:14:53 -0800 (PST), milk <milkm...@gmail.com>
wrote:

I'm on a topic modeling list
https://lists.cs.princeton.edu/mailman/listinfo/topic-models. There is a
lot of related research going on there. For performance reasons, nearly
all of the code is in either Java or C++, although there are Python and
R "wrappers" for much of the code. Most of the code I've seen go by is
open source and could probably be incorporated into SwiftRiver if it can
be somehow wrapped in PHP or Python. I don't know what's inside Techmeme
or any of the "commercial" journalism automation products, though.

I've got all of the R NLP code I could find (and the underlying C++ and
Java code) in the Social Media Analytics Research Toolkit if you want to
play with that. It's all open source, so if you (or SwiftRiver) see
something you like, just grab it. ;-)
--
http://twitter.com/znmeb http://borasky-research.net

"A mathematician is a device for turning coffee into theorems." -- Paul
Erdős

[Jon Gosier]

Cool project, Ed we'll look into those for integration for sure.

@Milk We have the beginnings of a meme discovery product we call SwiftMeme here - http://github.com/ushahidi/swiftmeme.  It does some of what you're referring to but it's pretty rough at the moment...prototype stage.

Another example is the GlobalHealthHub.org project which is an aggregator that is powered by Swift APIs to essentially offer a techmeme-like site for global public health.  I'd be happy to put them in touch with you as I know they'd be keen to share lessons learned and some of their code.

--

Jonathan D. Gosier

About.me | GosDot.com | Swiftly.org | @jongos

ph | 315.889.1746 

skype | j.gosier

--
You received this message because you are subscribed to the Google Groups "SwiftRiver" group.
To post to this group, send email to swift...@googlegroups.com.
To unsubscribe from this group, send email to swiftriver+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/swiftriver?hl=en.

[Victor Miclovich]

definitely agree :)