The Research On Natural Language Information Hiding Based On Synonymy Substitution

With the development and popularization of Internet, electronic documents have become one of the chief means of communication. Researches on information hiding have been developed. Text information hiding embeds some secret messages into digital documents, and the third party can not be aware of the existences of secret messages. Text information hiding techniques mainly include steganography em-ployed for secret communication, and digital watermarking which is used for copy-right protection. At present, text information hiding has become one of the hot spots in the field of information security.This paper is mainly about a new technique of text information hiding based on natural language processing techniques, that is natural language information hiding. This technique, using natural language processing, changes the syntax, semantic at-tributes of the original text to embed hidden information, while preserving the mean-ing of the original text as much as possible. This paper includes the following con-tents:Firstly, the concept, characteristic and basic model of information hiding are presented in this paper. This paper also introduces the existing text-information-hiding techniques. We analyze the present algorithms based on syn-onymy substitution for Chinese text in detail, and summarize their mainly existing problems.Secondly, to deal with the disadvantages of current algorithms based on synon-ymy substitution for Chinese text, this paper proposes an improved steganographic algorithm based on synonymy substitution for Chinese text. First, the algorithm clas-sifies the synonymy sets with HowNet. Then for the non-totally interchangeable syn-onymy sets, obtains the collocational words of the synonymy from the context by anlyzing the dependency relationships. And then determines whether to do the sub-stitution according to the collocational words.Lastly, this paper implements a text watermarking system using the improved algorithm. This system consists of copyright-marking model and copyright-detecting model. Copyright-marking model assigns a disparate watermark and a disparate key for each user, and chooses the sentences of digital document which can be employed for synonymy substitution, and then embeds the watermark and checks codes using quadratic residue theory. The copyright-detecting model employs the saved water- marking sentences and saved watermark of user to do the copyright detection for each user. This watermarking system is an effective attempt for the application of natural language information hiding theory.