1. Download data set
http://people.csail.mit.edu/irennie/20Newsgroups/
Download to hadoop home
2. Go to HADOOP_HOME
$> cd /home/alloe/hadoop-0.18.0
$> tar xvfz 20news-18828.tar.gz
$> bin/hadoop dfs -put 20news-18828 20newsInput
$> bin/hadoop jar apache-mahout-core-0.1-dev.jar org.apache.mahout.classifier.cbayes.CBayesDriver 20newsInput/alt.atheism 20newsOutput
