Chang-Po Chiang

The web shell attacks (WebShells) have long been a source of persistent annoyance for administrators. They have become a major security concern in cloud computing environments since the scalability and distributed nature of the cloud services could intensify the potential risks and impacts of such attacks. In response, researchers have proposed numerous strategies to shield assets from WebShell intrusions. Consequently, this study proposes a method that utilizes the BPE (Byte Pair Encoding) for tokenization, the CodeBERT for extracting the word embedding vector of a given source code piece, and a deep model (GRU or Bidirectional GRU) for determining whether the code contains a WebShell. This architecture is designed to detect the presence of WebShells in PHP code through analysis of the source code. Our experimental results indicate that, the proposed method with GRU achieves the best performance, with an accuracy of 99.72%, a precision of 99.36%, and an F1-score of 99.36%. Furthermore, it outperforms the methods proposed in the prior related studies as presented in the paper.

The menace of fake news has gained attention through the trend of the convenience of information-sharing in the cyber world. The information-sharing lifestyle has penetrated people's cyber routines. This study adopts a victimology perspective to explore the characteristics of the audience that is most likely to be influenced by fake news in social media through secondary data quantitative research. The results show that demographic characteristics, gender, and education level affect the likelihood of receiving fake news. When people use more social media platforms and have greater engagement, such as more sharing, reposting, and replying on social media platforms, there is a greater chance of receiving fake news.

This paper presents a novel reversible data hiding (RDH) technique based on pixel value ordering (PVO) and median preserving (MP) for improving image quality. In the proposed method, an adaptive block-wise embedding mechanism is applied. In PVO-based RDH, embedding data into a smooth block is always leading to a better image quality. For each block, the complexity is first calculated using surrounding blocks and then set a threshold in order to measure whether this block is a smooth or a complex block. The smooth blocks are embedded with secret data, and the complex blocks are skipped based on the threshold value. Therefore, this mechanism better exploits smooth blocks for data embedding, resulting in improved image quality. Experimental results demonstrate that the proposed method achieves a higher PSNR value than the previous approach.