udesign and
implementation of a system that categorizes email messages
ugoal: identify as
many unsolicited email messages (³spam²) as possible
vavoid false positives
(valid messages mis-classified as spam)
umethods
uuse content-based and
possibly usage-based techniques rather than explicit rules to filter
out spam
vBayesian
networks
vCollaborative
filtering
uBackground: AI, SE;
possibly in combination with knowlets
uFunding: None