Nowadays e-mail spam is not a novelty, but it is still an important rising problem with a big economic impact in society. Fortunately, there are different approaches able to automatically detect and remove most of those messages, and the best-known ones are based on machine learning techniques, such as Naïve Bayes classifiers and Support Vector Machines. However, there are several different models of Naïve Bayes filters, something the spam literature does not always acknowledge. In this chapter, we present and compare seven different versions of Naïve Bayes classifiers, the well-known linear Support Vector Machine and a new method based on the Minimum Description Length principle. Furthermore, we have conducted an empirical experiment on six public and real non-encoded datasets. The results indicate that the proposed filter is easy to implement, incrementally updateable and clearly outperforms the state-of-the-art spam filters. © 2012 Springer-Verlag Berlin Heidelberg.394 199214Almeida, T., Yamakami, A., Content-based spam filtering (2010) Proceedings of the 23rd IEEE International Joint Conference on Neural Networks, pp. 1-7. , Barcelona, SpainAlmeida, T., Yamakami, A., Almeida, J., Evaluation of approaches for dimensionality reduction applied with naive Bayes anti-spam filters (2009) Proceedings of the 8th IEEE International Conference on Machine Learning and Applications, pp. 517-522. , Miami, FL, USAAlmeida, T., Yamakami, A., Almeida, J., Filtering spams using the minimum description length principle (2010) Proceedings of the 25th ACM Symposium on Applied Computing, pp. 1856-1860. , Sierre, SwitzerlandAlmeida, T., Yamakami, A., Almeida, J., Probabilistic anti-spam filtering with dimensionality reduction (2010) Proceedings of the 25th ACM Symposium on Applied Computing, pp. 1804-1808. , Sierre, SwitzerlandAlmeida, T., Almeida, J., Yamakami, A., How the dimensionality reduction affects the accuracy of Naive-Bayes classifiers (2011) Journal of Internet Services and Applications, 1 (3), pp. 183-200Androutsopoulos, I., Paliouras, G., Michelakis, E., Learning to filter unsolicited commercial e-mail (2004) Technical Report 2004/2 National Centre for Scientific Research "demokritos", , Athens, GreeceBaldi, P., Brunak, S., Chauvin, Y., Andersen, C.A.F., Nielsen, H., Assessing the accuracy of prediction algorithms for classification: An overview (2000) Bioinformatics, 16 (5), pp. 412-424Barron, A., Rissanen, J., Yu, B., The minimum description length principle in coding and modeling (1998) IEEE Transactions on Information Theory, 44 (6), pp. 2743-2760. , PII S0018944898052845Cormack, G., Email spam filtering: A systematic review (2008) Foundations and Trends in Information Retrieval, 1 (4), pp. 335-455Drucker, H., Wu, D., Vapnik, V., Support vector machines for spam categorization (1999) IEEE Transactions on Neural Networks, 10 (5), pp. 1048-1054Forman, G., Scholz, M., Rajaram, S., Feature shaping for linear SVM classifiers (2009) Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 299-308. , Paris, FranceGrünwald, P., A tutorial introduction to the minimum description length principle (2005) Advances in Minimum Description Length: Theory and Applications, pp. 3-81. , Grünwald, P., Myung, I., Pitt, M. (eds.). MIT PressGomez Hidalgo, J.M., Evaluating cost-sensitive unsolicited bulk email categorization (2002) Proceedings of the ACM Symposium on Applied Computing, pp. 615-620John, G., Langley, P., Estimating continuous distributions in bayesian classifiers (1995) Proceedings of the 11st International Conference on Uncertainty in Artificial Intelligence, pp. 338-345. , Montreal, CanadaKolcz, A., Alspector, J., SVM-based filtering of e-mail spam with content-specific misclassification costs (2001) Proceedings of the 1st International Conference on Data Mining, pp. 1-14. , San Jose, CA, USALemire, D., Scale and translation invariant collaborative filtering systems (2005) Information Retrieval, 8 (1), pp. 129-150. , DOI 10.1023/B:INRT.0000048492.50961.a6Liu, S., Cui, K., Applications of support vector machine based on boolean kernel to spam filtering (2009) Modern Applied Science, 3 (10), pp. 27-31Losada, D., Azzopardi, L., Assessing multivariate Bernoulli models for information retrieval (2008) ACM Transactions on Information Systems, 26 (3), pp. 1-46Marsono, M., El-Kharashi, N., Gebali, F., Targeting spam control on middleboxes: Spam detection based on layer-3 e-mail content classification (2009) Computer Networks, 53 (6), pp. 835-848Matthews, B., Comparison of the predicted and observed secondary structure of T4 phage Lysozyme (1975) Biochimica et Biophysica Acta, 405 (2), pp. 442-451McCallum, A., Nigam, K., A comparison of event models for naive Bayes text classication (1998) Proceedings of the 15th AAAIWorkshop on Learning for Text Categorization, pp. 41-48. , Menlo Park, CA, USAMetsis, V., Androutsopoulos, I., Paliouras, G., Spam filtering with naive Bayes -which naive Bayes? (2006) Proceedings of the 3rd International Conference on Email and Anti-Spam, pp. 1-5. , Mountain View, CA, USARissanen, J., Modeling by shortest data description (1978) Automatica, 14, pp. 465-471Sahami, M., Dumais, S., Hecherman, D., Horvitz, E., A Bayesian approach to filtering junk e-mail (1998) Proceedings of the 15th National Conference on Artificial Intelligence, pp. 55-62. , Madison, WI, USASchneider, K.-M., On word frequency information and negative evidence in Naive Bayes text classification (2004) Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science), 3230, pp. 474-485. , Advances in Natural Languag Processing - 4th International Conference, EstAL 2004Sculley, D., Wachman, G.M., Relaxed online SVMs for spam filtering (2007) Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR'07, pp. 415-422. , DOI 10.1145/1277741.1277813, Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR'07Sculley, D., Wachman, G., Brodley, C., Spam filtering using inexact string matching in explicit feature space with on-line linear classifiers (2006) Proceedings of the 15th Text REtrieval Conference, pp. 1-10. , Gaithersburg, MD, USASiefkes, C., Assis, F., Chhabra, S., Yerazunis, W.S., Combining winnow and orthogonal sparse bigrams for incremental spam filtering (2004) Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3202, pp. 410-421Song, Y., Kolcz, A., Gilez, C., Better naive Bayes classification for high-precision spam detection (2009) Software - Practice and Experience, 39 (11), pp. 1003-1024Vapnik, V.N., (1995) The Nature of Statistical Learning Theory, , Springer, New YorkZadeh, L., Fuzzy sets (1965) Information and Control, 8 (3), pp. 338-353Zhang, L.E., Zhu, J., Yao, T., An evaluation of statistical spam filtering techniques (2004) ACM Transactions on Asian Language Information Processing, 3 (4), pp. 243-269. , DOI 10.1145/1039621.1039625, Computer Processing of Oriental Languages