Phishing involves manipulating individuals into revealing private data, e.g., user IDs, bank details, and passwords. The observed surge in fraud is related to increased deception, impersonation, and advanced online attacks. Thus, effective phishing detection methods are required to mitigate escalating global phishing threats. Existing methods (e.g., heuristics-based, signature-based, and visual similarity-based methods) attempt to detect phishing sites, and machine learning (ML) and deep learning (DL) methods are effective in the cybersecurity context in terms of learning from data, offering insights, and forecasting. However, independent ML algorithms are limited when handling complex data, and DL techniques surpass traditional ML methods in terms of performance but require more data and time. To tackle these challenges, we present EnLeM, an ensemble learning model designed specifically for phishing website detection. EnLeM brings together three well-known machine learning classifiers—decision tree, random forest, and k-nearest neighbor—using a hard voting mechanism, and further strengthens efficiency with Mutual Information–based feature selection. When tested on the UCI phishing dataset, EnLeM delivered strong results, reaching 97.21% accuracy and a 97.51% F1-score. Compared to individual ML classifiers, it consistently performed better, and it also proved more efficient than deep learning models such as CNN and LSTM. Notably, EnLeM maintained stable accuracy across different feature subsets while cutting execution time by roughly 13%. By striking a balance between accuracy, speed, and interpretability, EnLeM stands out as a practical and scalable solution for real-time phishing detection without the heavy resource demands of deep learning approaches.

Graph Neural Networks (GNNs) have demonstrated superior performance in node classification tasks across diverse applications. However, their vulnerability to adversarial attacks, where minor perturbations can mislead model predictions, poses significant challenges. This study introduces GOttack, a novel adversarial attack framework that exploits the topological structure of graphs to undermine the integrity of GNN predictions systematically. By defining a topology-aware method to manipulate graph orbits, our approach generates adversarial modifications that are both subtle and effective, posing a severe test to the robustness of GNNs. We evaluate the efficacy of GOttack across multiple prominent GNN architectures using standard benchmark datasets. Our results show that GOttack outperforms existing state-of-the-art adversarial techniques and completes training in approximately 55% of the time required by the fastest competing model, achieving the highest average misclassification rate in 155 tasks. This work not only sheds light on the susceptibility of GNNs to structured adversarial attacks but also shows that certain topological patterns may play a significant role in the underlying robustness of the GNNs. Our Python implementation is shared at https://github.com/cakcora/GOttack.

Social networks have become a popular way for internet surfers to interact with friends and family members, reading news, and also discuss events. Users spend more time on well-known social platforms (e.g., Facebook, Twitter, etc.) storing and sharing their personal information. This information together with the opportunity of contacting thousands of users attract the interest of malicious users. They exploit the implicit trust relationships between users in order to achieve their malicious aims, for example, create malicious links within the posts/tweets, spread fake news, send out unsolicited messages to legitimate users, etc. In this paper, we investigate the nature of spam users on Twitter with the goal to improve existing spam detection mechanisms. For detecting Twitter spammers, we make use of several new features, which are more effective and robust than existing used features (e.g., number of followings/followers, etc.). We evaluated the proposed set of features by exploiting very popular machine learning classification algorithms, namely k-Nearest Neighbor (k-NN), Decision Tree (DT), Naive Bayesian (NB), Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), and eXtreme Gradient Boosting (XG-Boost). The performance of these classifiers are evaluated and compared based on different evaluation metrics. We compared the performance of our proposed approach with four latest state of art approaches. The experimental results show that the proposed set of features gives better performance than existing state of art approaches.