10-04-2017, 09:02 PM
Hardware-Enhanced Association Rule Mining
with Hashing and Pipelining
For the purpose of implementing Apriori-based association rule mining in hardware, candidate itemsets and a database has to be loaded into the hardware. As the capacity of the hardware architecture is fixed, the items are loaded separately into the hardware if the number of candidate itemsets is larger than the hardware capacity. the number of candidate itemsets multiplied by the number of items in the database gives the complexity of the steps needed to load the itemsets into the hardware. the pipeline methodology in the HAPPI architecture is used to compare the database and itemsets and collecting useful information.
INTRODUCTION
Data mining technology is now used in a wide variety of fields . To list a few examples-analysis of customer transaction records, web site logs, credit card purchase information etc all need data mining. Information such as customer behaviorwill be available for the business managera and researchers. it is important to develop more efficient algorithms to extract knowledge from the data when the amount of data increases. parallel computing schemes has to be employed for more efficiency and when the volume of data increases at a much faster rate than the CPU speeds. But as the as the number of the parallel nodes grows, the performance cannot improve linearly. hardware
devices are hence used sometimes to accomplish mining tasks.
The HAPPI architecture includes a systolic array, a trimming filter, a hash table filter . The transactions when input into this architecture gives frequent patterns at the output.
Full report pdf:
http://selab.iecs.fcu.edu.tw/wiki/images...lining.pdf