Apriori algorithm for discovering frequent itemsets for mining Boolean association rules.
Motivation : http://cse.iitkgp.ac.in/~bivasm/uc_notes/07apriori.pdf
Original Paper :
Rakesh Agrawal and Ramakrishnan Srikant Fast algorithms for mining association rules in large databases. Proceedings of the 20th International Conference on Very Large Data Bases, VLDB, pages 487-499, Santiago, Chile, September 1994.
##Usage
The algorithm can be executed with (Both minimum support and minimum confidence lie between [0, 1]):
python apriori.py <data_set> <minimum_support> <minimum_confidence>
Example:
python apriori.py datasets/retail.csv 0.3 0.6
##Dataset:
retail.dat contains the (anonymized) retail market basket data from an anonymous Belgian retail store(Source: http://fimi.ua.ac.be/data/).
Additionally, retail.dat was converted into retail.csv using dat2csv.py provided in the repository.
##To-Do:
- Parallel Partition method for speeding up mining process (Reference : http://ieeexplore.ieee.org/abstract/document/553164/)