It's now very clear that you exactly want:

(a) You want a classification existing in the big data domain (how the big data can be classified and interpreted as a whole).

(b) You want to run some type classification analysis with data that you are having (you want to cluster/interpret data that you are having ).

For (a) your starting points can be this ones:

http://datascience.berkeley.edu/what-is-big-data/

http://www.datasciencecentral.com/profiles/blogs/how-many-v-s-in-big-data-the-characteristics-that-define-big-data

For (b) you can check this:

MLBase, Apache Sparkm, H20, etc.

MOA is for stream mining, right?

What about tool for offline data (big data dataset)?