Determine associative mining packages

[jeff1evesque]

We need to determine an acceptable associative mining package. One example is apache spark's frequent pattern mining. If this (or another implementation is chosen), we'll need to incorporate this solution to our docker + puppet build. One reason to choose apache spark, is since it doesn't provide association rules, as well as probit regression.