Abstract: To identify the copyright of natural language text effectively, this paper presents a novel scheme to derive birthmark from a text. Since the birthmark is a unique and native characteristic of every text, a text with the same birthmark of another can be easily suspected as a copy. It proposes an application scenario for birthmark, in which birthmark can be used as an effective method to identify the copyright of text. The experiments show that birthmark successfully distinguishes non-copied files and has quite a good tolerance against meaning-preserving attacks.

Key words: birthmark technology, digital watermarking, copyright protection, information hiding

摘要： 将胎记技术应用于文本，提出了一种全新的自然语言文本版权保护方案。从文本中提取代表特征信息的胎记，并由此识别文本的副本。在与数字水印相似的应用场景下，可用于解释版权归属，为文本提供有效版权保护。实验结果表明该胎记技术具有良好的置信性和鲁棒性。

关键词: 胎记技术, 数字水印, 版权保护, 信息隐藏

CLC Number: 

• TP309