Phishing attacks are more than two-decade-old attacks that attackers use to steal passwords related to financial services. After the first reported incident in 1995, its impact keeps on increasing. Also, during COVID-19, due to the increase in digitization, there is an exponential increase in the number of victims of phishing attacks. Many deep learning and machine learning techniques are available to detect phishing attacks. However, most of the techniques did not use efficient optimization techniques. In this context, our proposed model used random forest-based techniques to select the best features, and then the Brown-Bear optimization algorithm (BBOA) was used to fine-tune the hyper-parameters of the convolutional neural network (CNN) model. To test our model, we used a dataset from Kaggle comprising 11,000+ websites. In addition to that, the dataset also consists of the 30 features that are extracted from the website uniform resource locator (URL). The target variable has two classes: “Safe” and “Phishing.” Due to the use of BBOA, our proposed model detects malicious URLs with an accuracy of 93% and a precision of 92%. In addition, comparing our model with standard techniques, such as GRU (Gated Recurrent Unit), LSTM (Long Short-Term Memory), RNN (Recurrent Neural Network), ANN (Artificial Neural Network), SVM (Support Vector Machine), and LR (Logistic Regression), presents the effectiveness of our proposed model. Also, the comparison with past literature showcases the contribution and novelty of our proposed model.

Phishing attacks seriously threaten information privacy and security within the Internet of Things (IoT) ecosystem. Numerous phishing attack detection solutions have been developed for IoT; however, many of these are either not optimally efficient or lack the lightweight characteristics needed for practical application. This paper proposes and optimizes a lightweight deep-learning model for phishing attack detection. Our model employs a two-fold optimization approach: first, it utilizes the analysis of the variance (ANOVA) F-test to select the optimal features for phishing detection, and second, it applies the Cuckoo Search algorithm to tune the hyperparameters (learning rate and dropout rate) of the deep learning model. Additionally, our model is trained in only five epochs, making it more lightweight than other deep learning (DL) and machine learning (ML) models. The proposed model achieved a phishing detection accuracy of 91%, with a precision of 92% for the ’normal’ class and 91% for the ‘attack’ class. Moreover, the model’s recall and F1-score are 91% for both classes. We also compared our approach with traditional DL/ML models and past literature, demonstrating that our model is more accurate. This study enhances the security of sensitive information and IoT devices by offering a novel and effective approach to phishing detection.

The rapidly evolving darknet enables a wide range of cybercrimes through anonymous and untraceable communication channels. Effective detection of clandestine darknet traffic is therefore critical yet immensely challenging. This research demonstrates how advanced machine learning and specialized deep learning techniques can significantly enhance darknet traffic analysis to strengthen cybersecurity. Combining diverse classifiers such as random forest and naïve Bayes with a novel spiking neural network architecture provides a robust foundation for identifying concealed threats. Evaluation on the CIC-Darknet2020 dataset establishes state-of-the-art results with 98% accuracy from the random forest model and 84.31% accuracy from the spiking neural network. This pioneering application of artificial intelligence advances the frontiers in analyzing the complex characteristics and behaviours of darknet communication. The proposed techniques lay the groundwork for improved threat intelligence, real-time monitoring, and resilient cyber defense systems against the evolving landscape of cyber threats.