Comparison of a SOM based sequence analysis system and naive Bayesian classifier for spam filtering

@inproceedings{Luo05, abstract = {The problem introduced by the unsolicited bulk emails, also known as "spam" generates a need for reliable anti-spam filters. In this paper, we design and compare the performance of a newly designed SOM based sequence analysis (SBSA) system for the spam filtering task. The system is based on a SOM based sequential data representation combined with a kNN classifier designed to make use of word sequence information. We compare this system with the traditional baseline method naive Bayesian filter. Three different cost scenarios and suitable cost-sensitive measurements are employed. The results show that the SBSA system is superior to the naive Bayesian filter, particularly when the misclassification cost for non-spam message is high.}, author = {Luo, Xiao and Zincir-Heywood, Nur A. }, booktitle = {Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on}, citeulike-article-id = {3042401}, doi = {http://dx.doi.org/10.1109/IJCNN.2005.1556308}, journal = {Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on}, keywords = {cas\_chapter, email, hierarchical\_som, security, som, spam}, pages = {2571--2576 vol. 4}, posted-at = {2008-07-25 11:19:50}, priority = {5}, title = {Comparison of a SOM based sequence analysis system and naive Bayesian classifier for spam filtering}, url = {http://dx.doi.org/10.1109/IJCNN.2005.1556308}, volume = {4}, year = {2005} }

TY - CONF ID - Luo05 L3 - citeulike-article-id:3042401 TI - Comparison of a SOM based sequence analysis system and naive Bayesian classifier for spam filtering T2 - Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on JF - Neural Networks, 2005. IJCNN '05. Proceedings. 2005 IEEE International Joint Conference on VL - 4 SP - 2571 EP - 2576 vol. 4 N2 - The problem introduced by the unsolicited bulk emails, also known as "spam" generates a need for reliable anti-spam filters. In this paper, we design and compare the performance of a newly designed SOM based sequence analysis (SBSA) system for the spam filtering task. The system is based on a SOM based sequential data representation combined with a kNN classifier designed to make use of word sequence information. We compare this system with the traditional baseline method naive Bayesian filter. Three different cost scenarios and suitable cost-sensitive measurements are employed. The results show that the SBSA system is superior to the naive Bayesian filter, particularly when the misclassification cost for non-spam message is high. KW - cas_chapter KW - email KW - hierarchical_som KW - security KW - som KW - spam AU - Luo, Xiao AU - Zincir-Heywood, Nur PY - 2005/// UR - http://dx.doi.org/10.1109/IJCNN.2005.1556308 ER -