No abstract

Protecting sensitive information transmitted via public channels is a significant issue faced by governments, militaries, organizations, and individuals. Steganography protects the secret information by concealing it in a transferred object such as video, audio, image, text, network, or DNA. As text uses low bandwidth, it is commonly used by Internet users in their daily activities, resulting a vast amount of text messages sent daily as social media posts and documents. Accordingly, text is the ideal object to be used in steganography, since hiding a secret message in a text makes it difficult for the attacker to detect the hidden message among the massive text content on the Internet. Language’s characteristics are utilized in text steganography. Despite the richness of the Arabic language in linguistic characteristics, only a few studies have been conducted in Arabic text steganography. To draw further attention to Arabic text steganography prospects, this paper reviews the classifications of these methods from its inception. For analysis, this paper presents a comprehensive study based on the key evaluation criteria (i.e., capacity, invisibility, robustness, and security). It opens new areas for further research based on the trends in this field.

The rapid growth of online communication has increased the demand for secure communication. Most government entities, healthcare providers, the legal sector, financial and banking, and other industries are vulnerable to information security issues. Text steganography is one way to protect secure communication by hiding secret messages in the cover text. Hiding a high amount of secret information without raising the attacker's suspicion is the main challenge in steganography. This paper proposes the Color and Spacing Normalization stego (CSNTSteg) model to resolve the low capacity and invisibility problem on text steganography. CSNTSteg consists of two stages: the pre-embedding stage, which achieves high capacity by utilizing RGB coding and character spacing. It is designed to increase the number of bits per location and usable characters. Besides, it applies the Huffman coding technique to compress the secret message to add more capacity enhancement. The second stage is color and spacing normalization, which accomplishes high invisibility by normalizing the RGB coding and character spacing of the cover and stego text. CSNTSteg overcomes the color differences issue between the cover and stego texts regardless of the color of the cover text. To assess the quality of CSNTSteg, the experimental results are compared with existing works. CSNTSteg shows superior capacity over the existing studies with a percentage of 98.85%. CSNTSteg also achieves high invisibility by reducing the color differences with a percentage of 4.7% and 5.07% for black and colored cover text, respectively. Furthermore, CSNTSteg improves robustness by 94.22% by reducing the distortion in stego text. Overall, the CSNTSteg model embeds a high capacity of secret data while maintaining invisibility and security, offering a new perspective on text steganography to protect against visual and statistical attack issues.

No abstract

Cloud computing not only provides high availability on elastic resources, scalable, and cost-efficient. The platform is also widely used in information technology (IT) to support technology infrastructure and services. However, due to the complex environment and scalability of services, one of the highest security issues is malware attacks, where some of the antivirus scanner unable to detect metamorphic malware or encrypted malware where these kinds of malware able to bypass some traditional protection solution. This is why a high recognition rate and good precision detection are important to eliminate a high false-positive rate. Machine Learning (ML) classifiers are a critical role in artificial intelligent-system. However, machine learning will require to learn from the high amplitude of input data; classify then only able to generate a reliable model with a high detection rate. The objective of this work is to study and performs detection based on dynamic malware analysis and classification is through the WEKA classifier and Random Forest Jupyter Notebook. There are three classifiers chosen in this work, which are Random Forest, J-48, and Naive Bayes with 10-folds validation from the WEKA tool and another additional classifier from Random Forest -Jupyter Notebook to substantiate the accuracy.