Abstract: The sophistication of phishing assaults is rising, making it challenging to identify them using conventional means. Consequently, there will be a growing need for more advanced techniques to recognise and thwart such attacks. We introduce a hybrid deep learning strategy for phishing attack detection in this research. The proposed method of phishing website detection can be done by combining convolutional neural networks (CNN) and recurrent neural networks (RNN), two alternative deep learning models. The RNN is used to identify temporal dependencies in the data, while features have been determined from the unprocessed information using CNN. The hybrid model is highly accurate at identifying phishing assaults since it is trained on a vast dataset of authentic and phishing websites. We demonstrate that the suggested algorithm outperforms leading-edge methods by comparing its performance to those. Overall, our suggested method offers a reliable defence against phishing assaults and can be applied to increase the security of online systems.
Keywords: Legitimate, Phishing, Cyber Security, Deep learning, Feature Extraction, Websites, CNN, RNN.
|
DOI:
10.17148/IJARCCE.2023.124205
