The information content values come from WordNet::Similarity. These values are taken from the sense-tagged SemCor corpus and computed using the method described by Resnik 1995 :

You can see a bit about how we use information content in the source code documentation of WordNet::Similarity :

WordNet::Similarity does allow you to use other corpora to compute information content too.

I hope this helps.

Good luck!

On Tue, Nov 13, 2012 at 10:39 AM, Nachiket Kamat <> wrote:
Hello Ted,

To calculate IC(a) we 1st need to calculate p(a)=freq/N

Could you tell me from where does the program get value for freq? Is it corpus specific or does it pick this value from WordNet tool.


Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
senserelate-users mailing list