Given any sequence of letters, this program guesses what they might stand for (it finds potential backronyms).

It does this by finding all matches for the sequence of letters from a set of categories and a set of templates.

A category is a set of phrases (one or more words) that are interchangeable in some context, and a template is a string to be interpolated with category words.

After generating all the candidate backronyms, it sorts them with a scoring heuristic and displays the top result (or the entire list).

If we have the following two categories:

FRUIT: apple, banana, blueberry, papaya
DRINK: juice, milkshake, shake, smoothie

And the following templates:

FRUIT DRINK
FRUIT and FRUIT DRINK

And we query the engine with the acronym bps, we would get the following matches:

banana and papaya shake
banana and papaya smoothie
blueberry and papaya shake
blueberry and papaya smoothie

• Python 3.x

Download this repository with the following git command:

git clone https://github.com/benfrankel/backronym

Navigate into the downloaded folder and run python repl.py to interact with the engine.

This project is licensed under the Apache 2.0 license, so you are free to use, distribute, and modify it.