Performance Analysis on Mail Classification with Multilayer Perceptron (MLP)

Abstract

Spam is a key problem in electronic communication, including large-scale email systems. Classification of spam emails is a significant operation in email system. The efficiency of this process is determined by many factors such as number of features, feature selection techniques, representation of symbols and classifier. This paper focuseson email classifier with Multilayer Perceptron(MLP) approach for spam and ham mails classification. The systemis used termfrequency and inverse document frequency (tf-idf) and fisher score feature selection methods at preprocessing. These methods allow selecting relevant features and adding benefit in terms of improvisation in accuracy and reduced time complexity to email classification system.