The flooding of spam emails in email servers is an arm-racing issue. Even until today, filtering spam from email messages has become an ongoing work by researchers. Among all the methods proposed, methods that use machine-learning algorithms have achieved more success in spam filtering; unfortunately face a high dimensionality of features space after pre-processing and become a big hurdle for the classifier. Besides, the excessive number of features also can degrade the classification results. Thus, in this research, two stages of feature selection based on Taguchi methods were proposed to reduce the high dimensionality of features and obtain a good classification result for spam filtering