config

# Copy your data-set file into 'dataset' directory and provide the filename here.
MBA.txt
# Input directory path in hdfs; Make sure the directory is not already created. Existing directory in hdfs with the same name you provide here will be overwritten. If your hdfs home directory path is '/user/username' then you should provide below the full path as '/user/username/input-dir-name'. For users working under '/' in hdfs just write '/input-dir-name'.
/mba-input
# Output directory path in hdfs. Make sure you don't have this directory already created in hdfs. Exisitng directory with this same name will be overwritten.
/mba-output
# Full Path of output folder in your local file system. If the ouput directory you provided above exists at the provided path below then it will be overwritten.
/home/hadoop_pranit
# Set the default delimiter for your input data-set. If it's a .csv file then the delimiter should be ',' or if contents are space separated for each transaction then denote space delimiter within qoutes as " ".
" "
# Minimum support value. For 20% provide the value as 0.2
0.1
# Minimum confidence value. For 60% provide the value as 0.6
0.8
# Maximum number of passes for finding all frequent item-sets. This value will be used to terminate the algorithm if it doesn't converge before running for the maximum specified number of passes.
10
# Set to a value of 1 to filter the Rules generated by only postive lift, i.e., any Rules having a lift <= 1.0 will be discared. Set to a value of 0 to turn off this filter.
1