# Grab newsfeeds that you define. We use caching to make sure that they are not fetched too frequently. # Stuff them into a MySQL database. We have to be a little tricky here so we are not storing duplicate articles, even if the same article appears in several feeds. MySQL came to the rescue here. # Next, we call on the Content Analysis web service from Yahoo! to extract the keywords from the article titles and descriptions. The keywords that we get back are inserted into the database and associated with the appropriate articles. # Everything else is gravy. We pull out the popular keyword tags and use a silly scaling function to assign css according to how often the tag shows up. # Clicking a tag's link will display all the article abstracts for that particular tag as well as display the other tags associated with each article.

NOTE: forget about yahoo developer keyword analysis, as we can do it without them.

On 18/10/05, Alex Polite m4@polite.se wrote:

On Tue, Oct 18, 2005 at 03:22:42PM +0200, Jad Madi wrote:

...

Hi, any idea how to create tag clouds for aggregated feeds?

What's a tag cloud?

-- Alex Polite http://flosspick.org -- [ Drupal support list | http://lists.drupal.org/ ]

-- Regards Jad madi Blog http://EasyHTTP.com/jad/ Web standards Planet http://W3planet.net/

No it's not, maybe I have to explain what i'm trying to do further.

What I want is to SCAN 'description' field in Feeds, and get the most users keywords inside the description field.

so it will be something like LIVE tags.

in other words, I don't want to add taxonomy/tags, I just want to scan and use the keyword they are using in their feeds as tags