**************************************************************************************
README.txt

Email: nu-minebench AT ece DOT northwestern DOT edu
Contents: Explains how to setup and execute NU-MineBench

**************************************************************************************
-------------------------------------------------------------------
TO DOWNLOAD:
-------------------------------------------------------------------
Go to: http://cucis.ece.northwestern.edu/projects/DMS/MineBench.html



-------------------------------------------------------------------
TO INSTALL:
-------------------------------------------------------------------

tar -xvzf NU-MineBench-1.0.tar.gz


-------------------------------------------------------------------
FILE DIRECTORY DESCRIPTION:
-------------------------------------------------------------------

NU-MineBench is a collection of data mining applications. Currently 
there are 8 applications in the suite. 

Here's the files structure. Lets call $DMHOME as the home for our data 
mining suite NU-MineBench-1.0. The following is the file organization
within the root directory $DMHOME.

* README.txt - this file

* commandLine.txt - sample of the command line for executing each application. 
   You can similary extend it to other options/data sets. 
   This script uses $DMHOME as the home dir for applications. 
   So please "setenv" or "set" or "export" this variable before 
   trying out the options.

* src - containing all the source files of the applications

* datasets - contains all the data sets for each application 
             (subdirectory structure is the same as src)




Within src, you will find the following:
* APR - apriori based association rule application (horizontal database)

* ECLAT - another association rule application (vertical database)

* Bayesian - A naive bayesian classifier application

* ScalParC - A decision tree based classification application

* birch - Hierarchical clustering application

* kmeans - Partitioning based clustering application

* kmeans also contains a fuzzy based clustering application (execute with option -f)

* hop - Density based clustering application used in astrophysics



Within datasets, you would find the relevant datasets for each application 
(follows the same directory as above).


-------------------------------------------------------------------
COMPILATION:
-------------------------------------------------------------------
APR:
cd $DMHOME/src/APR
make

ECLAT:
cd $DMHOME/src/ECLAT
make

Bayesian:
cd $DMHOME/src/Bayesian/bayes/src
make

ScalParC:
cd $DMHOME/src/ScalParC/
make

birch:
cd $DMHOME/src/birch
make

kmeans:
cd $DMHOME/src/kmeans
make example

hop:
cd $DMHOME/src/HOP
make



-------------------------------------------------------------------
EXECUTION:
-------------------------------------------------------------------

See commandLine.txt file for sample

NOTE: for any application, just typing the application name without any command 
line options would list the actual command line options that are available to the
user. Feel free to use all datasets provided.



Thanks for using NU-MineBench. 
For issues send email to nu-minebench AT ece DOT northwestern DOT edu

**************************************************************************************