This algorithm is written in C++. It will read the input file (text file containing all the transactions in the DB) and will output the results to a file (provided by user as the command argument).

make

./apriori <minsup_count> <K> <input_file> <output_file>

minsup_count: Minimum support count (minsup_percentage * number of transactions)
K: frequent K-itemset
input_file: transaction dataset 
output_file: final result file

Note: all the arguments are required.

The input file should have the following input. Each line represents a transaction, which each of the itemsets in the transactions are separated by a "space". Itemsets can be composed of any character set of strings.

Hint: If your format changes from what was discribed above, you can easily change only the "readFile" function the file.

In this code, speed and efficiency was the first goal. The algorithm is designed to provide the results as fast as possible considering exploiting only one code.

Work-in-progress: Using OpenMP features