While building our own versions of logistic regression, naive Bayes, and decision trees has provided a valuable opportunity to talk about the theory behind them, Clojure gives us several libraries for building classifiers. One of the better supported is the clj-ml library.
The clj-ml library is currently maintained by Josua Eckroth and is documented on his GitHub page at https://github.com/joshuaeckroth/clj-ml. The library provides Clojure interfaces for running linear regression described in the previous chapter, as well as classification with logistic regression, naive Bayes, decision trees, and other algorithms.
Note
The underlying implementation for most machine learning functionality in clj-ml is provided by the Java machine learning library Weka. Waikato Environment for Knowledge Analysis (Weka), an open source machine learning project released and maintained primarily by the Machine Learning Group at the University of Waikato, New Zealand (http://www.cs.waikato...