This is a set of utilities for ranking the fitness of source strings based on ngram statistics. The string "Sample" could be broken up into the following 3-character substrings:

• SAM
• AMP
• MPL
• PLE

These substrings can be compared to how frequently they appear in the English language, and a score can be assigned. This library includes tools to generate such frequency lists from provided strings, as well as a scorer to rank strings based on those frequencies. It also includes examples of using the scorer to attack substitution ciphers.

For more discussion of the basic technique and precomputed nGram files, see http://practicalcryptography.com/cryptanalysis/text-characterisation/quadgrams/ (not written by me)