This is a prototype to try out PEMA class in RevoPemaR package.
The purpose of this project is to show one possible way to utilize PEMA class through Apriori algorithm application.
PEMA stands for Parallel External Memory Algorithm, which does not require all of the data to be in memory at one time.
RevoPema Getting Started Guide
- Download MovieLens datasets and unzip them under
org-data
folder which is in your home directory. - Create a folder named
input
in your home directory. - Execute
data_conversion.R
. This creates a bunch of files ininput
folder. - Execute
PemaAprioriExec.R
, maybe first and later half separately.
- How to eliminate redundant rules from the result of Apriori.
- How to adjust Apriori computation when ids are sparsed integers.
- How to distribute Apriori algorithm execution.