SPAM TECHNOLOGIES TALK, Lueg, Nov. 12

[Erna Amerman]

Prof Christopher Lueg
Computer Sciences Corp. Professor of eBusiness at Charles Darwin
University
http://www-staff.it.uts.edu.au/~lueg/

will be visiting GSLIS on Friday Nov 12.
He will be giving a talk in rm126 at 1pm


If you would like to chat to him about this or his other research,
please
contact Michael Twidale (twidale at uiuc.edu)


Fighting Spam: A Different Call to Arms


Spam emails are now recognized as a major problem costing the community
billions of dollars per annum. One of the impacts of the spam flood is
the
ubiquitous deployment of technical anti-spam measures, in particular
spam-filtering mechanisms removing or tagging individual emails
classified
as spam. As a loosely coupled network, anti-spam measures have the
potential of influencing communication to an extent that is without
precedence in human communication. This talk is about how spam filters
actually accomplish what they appear to be doing. Technically, spam
filters classify email messages into "spam" and "non-spam". A closer
look
at the technology reveals significant differences between definitions of

spam operationalized in spam filters and those used in spam discourses.
Spam or, more precisely, unsolicited (commercial) email, is typically
defined in a user-referential way: unsolicited refers to the message
recipient and his or her attitude towards receiving the message.
Definitions of spam operationalized by spam filters, however, focus on
message characteristics, such as the occurrence and distribution of
terms
considered typical of spam. This means messages are classified as
"solicited" or "unsolicited" regardless of the recipient's attitude
towards those messages. The reason appears to be that "solicitedness" is

not an objective, measurable aspect of the documents under
investigation.
In this respect, spam filtering appears to resemble traditional
information filtering/retrieval where relevance of documents to an
information need is not an objective, measurable aspect either.
Considering the growing impact of spam filtering, the community needs to

address these issues in a systematic, theoretically grounded way.