Markov-chain simulation of @brianleroux. Can you tell the difference?! (probably!)
JavaScript
Switch branches/tags
Nothing to show
Pull request Compare This branch is 1 commit behind notmatt:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
data
views
.gitignore
LICENSE
README.md
fakebrian.js
indexer.js
markov.js
package.json
server.js
token.js
toktag.js
twitter.js

README.md

README

A preliminary markov-chain generator from an index built on @brianleroux's last few thousand tweets.

node fakebrian.js will dump you to a REPL where fb.genTweet() will show you what's up.

node app.js will launch a web app on :3000.

Features:

  • Occasionally funny.

Doesn't feature:

  • Configurability (e.g., other users)
  • Posting to @fakebrianleroux
  • feedback/reinforcement (Brian would/wouldn't say that!)

Markov chains in a nutshell: given a particular state, in this case the previous N words, determine the next word. This is done via simple frequency, e.g.:

 key : ["a", "fluffy"],
 words : ["kitten", "cloud", "kitten", "kitten"]
 
 next word: "kitten" (75%)  "cloud" (25%)