Text Information Hiding Based on Part of Speech Grammar

Method for text information hiding based on part of speech (PoS) grammar is proposed. A context free PoS grammar generates PoS strings, which are patterns of natural language sentences. Taking PoS strings as a transform domain of natural language, the method maps both secret message and cover text to the domain, then maps each PoS string to a number where sentence appears in the cover text by string matching, and sequence of the numbers is the secret key. Using the secret key, receiver extracts the secret message by syntactic analysis. Further more, a formula computing the capacity of hidden information is given. This method does not require any modification of text, meanwhile, it is a sound solution of the problem cover text is syntactic but not semantic that Wayner ´s scheme encountered.