Investigation Into Parameterisation of the FIMOTS Algorithm for Computational Efficiencies on FPM on Streams of Data

O’Loughlin, Rory (2021) Investigation Into Parameterisation of the FIMOTS Algorithm for Computational Efficiencies on FPM on Streams of Data. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (288kB) | Preview

Abstract

Frequent pattern mining (FPM) has long been a data science technique with applications in tendency analysis and association rules. Technological evolutions have meant that data is no available within streams and expectations exist for near real-time processing as opposed to the batch processing of static datasets from the past. This introduces complexity in how existing algorithms may maintain accurate results of frequent patterns as new transactions arrive to the stream and older ones become less important. Usually there is a trade-off between the accuracy of the results and the amount of processing and memory required to achieve those results. This paper investigates the FIMoTS algorithm for FPM on data streams and determines how the accuracy of the frequent patterns identified by it is impacted by the introduction of a parameter which reduces the amount of computation required. Depending on the datasets used, it has been found that a large reduction in the amount and time spent in computation can be achieved with no loss of precision and a minor (<1%) reduction in recall.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Clara Chan
Date Deposited:	11 Dec 2021 11:14
Last Modified:	11 Dec 2021 11:14
URI:	https://norma.ncirl.ie/id/eprint/5204

Actions (login required)

View Item