Deep Neural Network (Deep-NN) technique for detecting web phishing in transactional system

https://doi.org/10.36386/4f48v545

Phishing is a prevalent cybercrime in which attackers deceive internet clients into divulging confidential details, such as credit card or financial information, in business-related transactions. Web phishing involves creating fraudulent websites that closely resemble legitimate ones, misleading users into submitting confidential information. Cybercriminals increasingly exploit phishing attacks to conduct security breaches because existing phishing detection systems lack robust defensive strategies. This study explores Deep Neural Networks (Deep-NN) as advanced classifiers for optimizing phishing detection performance. The dataset, sourced from a web repository, underwent pre-processing to extract relevant features and was cleaned for accurate classification. The experiment utilized 10,000 phishing data instances with 10 selected attributes, resulting in a 10,000 x 10 data dimension for simulation. The implementation was conducted using Python’s machine learning libraries, MATLAB for simulation, and Microsoft Excel for data preparation. The Deep-NN system was evaluated using scientific experiments, achieving improved phishing detection performance. Key results include a mean squared error (MSE) of 1094. `6, indicating the average squared difference between the classified output and the expected target, and a regression (R) value of 0.99, demonstrating a strong correlation between the predicted and actual values. As well as a high similarity coefficient in the Jaccard index. These findings show significant improvement over existing models. The study recommends further exploration of supervised learning algorithms to filter legitimate web traffic from large datasets and optimize classification performance.