[Snowball-discuss] Question regarding medical stemmer

[Chao Pang]

Dear developers,

My name is Chao Pang and I am a PhD student working at University Medical
Center of Groningen. We developed a system where we can match data elements
between any of the two data sources, in which we are actually using the
Snowball stemmer to process the terms prior to the matching process.

However I found out couple of exceptions that the standard Snowball stemmer
could not properly stem, here are two examples, I also tried the same
examples at your online demo http://snowball.tartarus.org/demo.php, but )

*Example1. asymmetry V.S. asymmetric*
asymmetry -> asymmetri
asymmetric -> asymmetr

*Example2. placenta V.S. placental*
placenta -> placenta
placental -> placent

I am not sure if I missed anything but I am wondering if there are any ways
that these two sets of terms can be stemmed correctly?

Are there any snowball stemmers that are extended for the medical domain?

-- 
Cheers
Chao
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tartarus.org/mailman/private/snowball-discuss/attachments/20140701/c753e670/attachment.html>