Ontological semantics spam filters
Abstract
E-mail fraud has become very prevalent in cyberspace and is currently a major technique utilized by cyber criminals to swindle victims. E-mail fraud is a category of spam or unsolicited bulk e-mail [1]. Spam filter research has been very active in combating e-mail spam. Spam filter research ranges from statistical methods for text categorization to newer methods of defining user preference ontologies to classify incoming e-mails. Much of these methods have limitations or an upper bound where they can be bypassed by simply misspelling, manipulating, or rephrasing the text. The research proposed in this composition utilizes a new technique that uses a very powerful tool known as ontological semantics. Ontological semantics gives direct access to the texts meaning, which in turn will help accurately classify and categorize unsolicited bulk e-mails. This study will provide insight on less effective current spam filter techniques and discuss their limitations compared to the proposed method of an ontological spam filter.
Degree
M.A.
Advisors
Raskin, Purdue University.
Subject Area
Linguistics|Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.