A Big Data Text Coverless Information Hiding Based on Topic Distribution and TF-IDF

Coverless information hiding has become a hot topic in recent years. The existing steganalysis tools are invalidated due to coverless steganography without any modification the carrier. However, for text relatively low capacity, this paper proposed big data method based on LDA (latent Dirichlet allocation) distribution and keyword TF-IDF (term frequency-inverse document frequency). Firstly, sender receiver build codebook, including word segmentation, frequency features, model clustering. then shreds secret information, converts it into ID through keywords-index table, searches containing keywords. Secondly, searched is taken as index tag according features. At same time, random numbers introduced control order of information.