Clojure Naive Machine Learning

As promised, I’m starting to add some machine learning content. Since I found most resources on machine learning are in Python, I figured it might be of interest to some people to have examples in other languages.

Of course, the implementations I’m suggesting are naive in the sense that they are not optimized at all. On the other hand, nothing is hidden in a library or framework: I’m only using Clojure core library at this point.

So far, I have only implemented the k-nearest neighbours algorithm putting the k=1 case separately.

I added a way to laod the famous iris flower dataset easily in order to test those implementations and I’m hoping people running this will be amazed to get accuracies above 95% with such simple code.

For reference, the code is here and the project’s page is here

More fun coming soon, stay tuned :)