الفهرس | Only 14 pages are availabe for public view |
Abstract Spam detection techniques are still incapable of totally stopping and hindering e-mail spam attacks. Some techniques mainly based on NaIve Bayesian algorithms as well as other Machine Learning algorithms such as Boosting trees or Support Vector Machines (SVM) were developed and used with some success. Other machine learning including neural network, Deep belief networks, fuzzy set, artificial immune system, as well as rough sets decision tree, can be successfully employed to tackle various problems such as spam detection. However, the number of False Positives (FP) and False Negatives (FN) resulting from the application of various spam e-mail filters still remains too high and the problem of spam e-mail categorization cannot be solved completely from a practical viewpoint. The scope of this thesis is to assess spam detection systems and discuss some representation methods to provide inspiring examples that illustrate how Machine Learning (ML) are utilized to address these problems and how spam data can be characterized by machine learning in addition to identifying which machine learning is close to perfection in battling e-mail spam. |