2026-05-23T01:44:43.469Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:3:9140 2024-12-03T20:10:47Z 3

Training Logistic Regression Model by Enhanced Moth Flame Optimizer for Spam Email Classification, Chapter in LNDECT Lecture Notes on Data Engineering and Communications Technologies: ICCNCT 2022: Computer Networks and Inventive Communication Technologies, Springer, volume 141 2022 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/3/9140 https://link.springer.com/chapter/10.1007/978-981-19-3035-5_56 M. Salb L. Jovanovic M. Zivkovic E. Tuba A. El-sadai N. Bacanin Spam email is a massive issue that bothers and consumes receivers’ time and effort. Because of its effectiveness in identifying mail as wanted or unwanted, machine learning approaches have become a popular technique in spam detection. Current spam detection methods, on the other side, typically have low detection performance and are incapable of handling high-dimensional information easily. As a result, a unique spam detection approach that combines an improved moth flame optimization algorithm and a logistic regression classification model was proposed in this paper. The research evidence on two accessible datasets (CSDMC2010, Enron) indicates that the suggested methodology can tackle high-dimensional data due to its very powerful local and global search skills. The suggested technique was evaluated for spam detection accuracy to that of logistic regression, naive Bayes classifiers, and support vector machine, as well as the performance of earlier research’ that includes state-of-the-art approaches. In terms of classification performance, the suggested methodology outperforms the other spam detection algorithms examined in this work. bookPart Springer, Singapore 753 768 10.1007/978-981-19-3035-5_56 LNDECT Lecture Notes on Data Engineering and Communications Technologies: ICCNCT 2022: Computer Networks and Inventive Communication Technologies, volume 141