I am currently working on a Wordpress MU based platform for several professional weblogs, and one of the things I’m goofing around with is the phenomenon of tags, plugged into Wordpress in the shape of Ultimate Tag Warrior.

One of the coolest features in this nifty little program is the usage of Yahoo’s term extraction API, which gives the editor tag suggestions when writing a new item. Unfortunately, this tool seems to be very much English oriented. When jamming Dutch texts into it, it does come up with results but most of them are far from usable.

To overcome this problem I figured that if I would be able to translate the Dutch text to English, then send it to the term extraction tool, translate whatever comes out back to Dutch and present those terms to the editor as tag suggestions. So I have edited Ultimate Tag Warrior to use Babel Fish to do just that. And I have to say, even though there is still some jibberish among the results, my plan turned out to work.

The only problem now is that the Dutch language uses a lot of English words, most of the time specific terms in business or science. And since these terms are so specific, they are probably words you want to have included in your tag suggestions. So right now I’m looking for a way to combine the original results with the results generated through Babel Fish. Any suggestions of course are welcome!