[Corpora-List] Email address identification

[Ron Artstein]

Asked on behalf of a sociologist colleague:

Does anyone know of a tool that runs through a corpus of email 
correspondence and identifies different email addresses which 
belong to the same person?

Obviously this can't be done with 100% certainty, and the ideal 
tool for my colleague's purpose should err on the side of unifying 
addresses. For example, if the tool is uncertain whether the 
following two people are the same:

Andy Smith <a.smith at company.com>
"Smith, A. C." <acs37 at service.net>

Then it should assume the two email addresses belong to the same 
person (until there is evidence to the contrary).

Thanks, -Ron.

_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora