Journal of Cybersecurity and Information Management JCIM 2690-6775 2769-7851 10.54216/JCIM https://www.americaspg.com/journals/show/1096 2019 2019 Faculty of Artificial Intelligence, Egyptian Russian University (ERU), Cairo, Egypt Mahmoud A. Zaher Department of Computer and Information Systems Sadat Academy for Management Sciences (SAMS), Cairo, Egypt Nabil M. Eldakhly Phishing is a familiar kind of cyberattack in the present digital world. Phishing detection with maximum performance accuracy and minimum computational complexity is continuously a topic of much interest. A novel technology was established for improving the phishing detection rate and decreasing computational constraints recently. But, one solution has inadequate for addressing every problem due to attackers from cyberspace. Thus, the initial objective of this work is for analysing the performance of different deep learning (DL) techniques from detection phishing activity. This study introduces a novel Brain Storm Optimization with Long Short Term Memory Enabled Phishing Webpage Classification (BSOLSTM-PWC) for Cybersecurity. The proposed BSOLSTM-PWC technique enables to accomplish cybersecurity by the identification and classification of phishing webpages. To accomplish this, the BSOLSTM-PWC technique initially employs data pre-processing technique to transform the data into actual format. Besides, the BSOLSTM-PWC technique employs LSTM classifier for the identification and categorization of phishing webpages. Moreover, the BSO algorithm is utilized to appropriately adjust the hyperparameters involved in the LSTM model. For reporting the improved outcomes of the BSOLSTM-PWC method, a wide-ranging simulation analysis is made using benchmark dataset. The experimental outcomes reported the enhanced outcomes of the BSOLSTM-PWC method on existing methods. 2022 2022 20 30 10.54216/JCIM.090202 https://www.americaspg.com/articleinfo/2/show/1096