Here are a few little software projects I’ve developed, mostly hosted through my GitHub account:
Make-rFiat’s a bit like autotest for make-managed projects. It monitors a make task’s dependencies. Each time one of those dependencies changes, fiat executes the task, prints the results, and reports whether the process succeeded or failed. Ruby.
Install it as a gem:
$ gem install fiat
Generates a random sentence modeled off of textual input using a Markov model. Give it a book and it’ll attempt to write a sentence in the style of that author. Python.
Reads in a grammar and vocabulary defined by the user and prints out a compatible random sentence. Think MadLibs + context-free grammars. Written way back in 2007. Common Lisp.
Useful for doing TF-IDF document comparison with an absolute minimum of fuss. Exactly two methods, handy for small-scale or throwaway information retrieval projects. Not too concerned about efficiency, though. Python.
Probably most useful if you'd like a clean example of how TF-IDF conceptually operates. If you want to do some serious information retrieval in Python, you might find a more heavy-weight tool like Gensim to be more appropriate.