• DocumentCode
    1843907
  • Title

    Text Zero-Watermark Based on Chinese Edit Distance

  • Author

    Meng Yingjie ; Wang Xianlong ; Liu Wenjun ; Chen Wei

  • Author_Institution
    Sch. of Inf. Sci. & Technol., Lanzhou Univ., Lanzhou, China
  • fYear
    2013
  • fDate
    21-23 June 2013
  • Firstpage
    686
  • Lastpage
    689
  • Abstract
    In research of the copyright protection of digital text, according to the shortage of poor robustness and weakness of attack resistance for the existing text watermark, Chinese machine code and information gain strategy have been introduced into text zero-watermark scheme, which puts forward the text zero-watermark based on the Chinese Edit Distance, constructs the construction and detection process model, and designs the core algorithm. First, we introduce information gain to compute the weighting terms in paragraph to select the text feature. Then, we use Chinese machine code to express the text feature formally. Last, we compute the edit distance between the characteristic words in every paragraph and mark the commutation position to form an edit distance matrix to complete text zero-watermark construction. In detection stage, we analyze the edit distance and commutation position to reflect the distortion status. In order to validate effectiveness and feasibility of the scheme, simulation experiments are provided. The results indicated that the scheme has good robustness and strong ability against attack, especially, for the attack toward sentence type conversion and add & delete words, it has more obvious advantages.
  • Keywords
    copyright; feature extraction; natural language processing; text analysis; text editing; watermarking; Chinese edit distance; Chinese machine code; attack resistance robustness; construction process model; detection process model; digital text copyright protection; edit distance matrix; information gain strategy; sentence type conversion; text feature selection; text zero-watermark scheme; Algorithm design and analysis; Computers; Encryption; Feature extraction; Resistance; Robustness; Watermarking; Digital text; Edit Distance; Information Gain; zero-watermark;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computational and Information Sciences (ICCIS), 2013 Fifth International Conference on
  • Conference_Location
    Shiyang
  • Type

    conf

  • DOI
    10.1109/ICCIS.2013.186
  • Filename
    6643102