Classifying and clustering malicious advertisement uniform resource locators using deep learning

Abstract

Malicious online advertisement detection has attracted increasing attention in recent years in both academia and industry. The existing advertising blocking systems are vulnerable to the evolution of new attacks and can cause time latency issues by analyzing web content or querying remote servers. This article proposes a lightweight detection system for advertisement Uniform resource locators (URLs) detection, depending only on lexical-based features. Deep learning algorithms are used for online advertising classification. After optimizing the deep neural network architecture, our proposed approach can achieve satisfactory results with false negative rate as low as 1.31%. We also design a novel unsupervised method for data clustering. With the implementation of AutoEncoder for feature preprocessing and t-distributed stochastic neighbor embedding for clustering and visualization, our model outperforms other dimensionality reduction algorithms by generating clear clusterings for different URL families.

Modeling Correlation between Android Permissions Based on Threat and Protection Level Using Exploratory Factor Plane Analysis

“…The framework was also inadequate in demonstrating the correlation that exists among those features extracted; thus, it cannot identify emerging Android malware. Using t-distribution stochastic neighbor embedding (t-SNE) for visualization, Zhang et al [58] proposed a method for detecting malicious adverts in addresses of web pages using lexical-based features. Even though HTTPs provide authentication on the web browsers and pages to offer data secrecy [59], the study of Barrera et al [60] demonstrated that sophisticated malware leverage on permission-based security vulnerabilities to infect Android and web-based applications.…”