A comparatively efficient software for embedding secret information bits into natural language texts

A place to discuss the science of computers and programs, from algorithms to computability.

Formal proofs preferred.

Moderators: phlip, Moderators General, Prelates

mok-kong shen
Posts: 15
Joined: Mon Oct 31, 2016 10:19 am UTC

A comparatively efficient software for embedding secret information bits into natural language texts

Postby mok-kong shen » Sun Dec 11, 2016 10:48 am UTC

Years ago in another Internet forum I asked for software for comparatively efficiently embedding secret information bits into natural language texts, without however obtaining useful answers. I have just implemented a scheme that should roughly satisfy my condition. A Python code is available at: http://s13.zetaboards.com/Crypto/topic/9024439/1/. For comments and critiques I should be very grateful.

Derek
Posts: 2148
Joined: Wed Aug 18, 2010 4:15 am UTC

Re: A comparatively efficient software for embedding secret information bits into natural language texts

Postby Derek » Mon Dec 12, 2016 4:22 am UTC

I'm going to guess that your prologue encodes some message using your system?

mok-kong shen
Posts: 15
Joined: Mon Oct 31, 2016 10:19 am UTC

Re: A comparatively efficient software for embedding secret information bits into natural language texts

Postby mok-kong shen » Mon Dec 12, 2016 10:24 am UTC

No (that doesn't worth my efforts). On the other hand IMHO the software has fairly convincingly shown that any apparently very neutral text, be it emails, posts to forums, contents of webpages or what not, can contain secret information bits roughly at a rate of 1/2 to 1 bit per word.

elasto
Posts: 3102
Joined: Mon May 10, 2010 1:53 am UTC

Re: A comparatively efficient software for embedding secret information bits into natural language texts

Postby elasto » Sun Feb 26, 2017 10:41 pm UTC

(Apologies in advance if your code already does this)

It seems to me that it'd be useful to base it off a thesaurus rather than a dictionary - with each set of equivalent words divided into three groups representing zero-bit, one-bit and ignore-bit

- The user writes an entire paragraph of normal text
- All words that correspond to an incorrect bit get highlighted
- The user clicks on each incorrect word and is presented with a dropdown list of all synonyms for that word that would represent a correct-bit (or ignore-bit)

(At the risk of the text sounding rather convoluted and erudite, each thesaurus entry could be split into five groupings: 00, 01, 10, 11 and <ignore> to double the transmission rate)


Return to “Computer Science”

Who is online

Users browsing this forum: No registered users and 7 guests