Feature Selection and Classification of Email Spam Using Orthogonal Linear Jellyfish Swarm Optimizer

Spam emails are the cyber annoyances which causes the significant security attacks involves both financial as well as personal data. Detecting the new strains of the spam messages is challenging and it needs the efficient and reliable intelligent spam detection approach. This research proposes the Orthogonal Linear Design based Jellyfish Swarm Optimizer (OLJSO)-based feature selection approach for the email spam. Initially, UCI ML Repository dataset is collected and it is pre-processed by using tokenization, stop word removal, stemming and lemmatization. After pre-processing, the feature extraction is done by using Term Frequency-Inverse Document Frequency (TF-IDF) and the extracted features are selected by the utilization of OLJSO. After the feature selection process, the selected spam features are classified into two types such as spam and non-spam by the utilization of Artificial Neural Network (ANN). The effectiveness of the OLJSO is estimated by using various performance metrices and it achieves the accuracy of 99.09% and precision of 0.975 when compared to the existing methods such as Genetic Decision Tree Processing with Natural Language Processing (GDTPNLP).