| Title | Unsupervised Deep Learning for Text Steganalysis |
| Publication Type | Conference Paper |
| Year of Publication | 2020 |
| Authors | Xu, Yueyao |
| Conference Name | 2020 International Workshop on Electronic Communication and Artificial Intelligence (IWECAI) |
| Date Published | jun |
| Keywords | anoamly detection, composability, Conferences, Deep Learning, learning (artificial intelligence), Metrics, Multimedia communication, privacy, pubcrawl, security, steganography, steganography detection, text steganalysis |
| Abstract | Text steganography aims to embed hidden messages in text information while the goal of text steganalysis is to identify the existence of hidden information or further uncover the embedded message from the text. Steganalysis has received significant attention recently for the security and privacy purpose. In this paper, we develop unsupervised learning approaches for text steganalysis. In particular, two detection models based on deep learning have been proposed to detect hidden information that may be embedded in text from a global and a local perspective. Extensive studies have been carried out on the Chinese poetry text steganography datasets. It is seen that the proposed models show strong empirical performance in steganographic text detection. |
| DOI | 10.1109/IWECAI50956.2020.00030 |
| Citation Key | xu_unsupervised_2020 |
Unsupervised Deep Learning for Text Steganalysis