期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2010
卷号：1
期号：2
出版社：Ayushmaan Technologies
摘要：In this paper, we have proposed an efficient approach for the reduction of significant attributes from the spambase warehouses for identifying spam email using forward selection method and have performed the classification of spam e-mail using data mining techniques. The data used in this paper is collected from UCI Machine Learning Repository Spambase dataset. The dataset consist of 4601 records which have 58 attributes and after applying Correlation–based Feature Selection methods the original attributes was reduced to 22, 16 and 8 potential attributes. We have investigated four data mining techniques such as J48, BayesNet, OneR and Classification via Clustering. The results shows that BayesNet have much better performance than other three methods and it is also observed that using feature reduction method the performance of BayesNet, OneR and classification via clustering a notable improvement in their classification.
