Skip to main navigation Skip to search Skip to main content

Synthesization of High-Utility Patterns in Parallel Computing

  • Jerry Chun Wei Lin
  • , Yuanfa Li
  • , Matin Pirouz
  • , Linlin Tang
  • , Miroslav Voznak
  • , Lukas Sevcik

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

High utility pattern mining (HUPM) has become a key issue in knowledge discovery since it provides retailers and managers with useful information for making decisions efficiently. However, previous studies most focused on mining the high-utility patterns (HUPs) from a single database. In this paper, we present a framework to incorporate the weighted model for parallel synthesis of the discovered HUPs from various databases. The pre-large concept was also used as a buffer here in order to provide more prospective HUPs, thus providing higher accuracy of the synthesized patterns. From our experiments, the developed model exceeds existing works, in particular the designed model has increased precision and recall on knowledge synthesization compared to the previous works.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE/ACM 23rd International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2019
EditorsFloriano De Rango, Carlos Tavares Calafate, Miroslav Voznak, Alfredo Garro, Mauro Tropea
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728129235
DOIs
StatePublished - Oct 2019
Externally publishedYes
Event23rd IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2019 - Cosenza, Italy
Duration: 8 Oct 20199 Oct 2019

Publication series

NameProceedings - 2019 IEEE/ACM 23rd International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2019

Conference

Conference23rd IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications, DS-RT 2019
Country/TerritoryItaly
CityCosenza
Period8/10/199/10/19

Keywords

  • parallel computing
  • pre-large concept
  • synthesization
  • utility patterns

Fingerprint

Dive into the research topics of 'Synthesization of High-Utility Patterns in Parallel Computing'. Together they form a unique fingerprint.

Cite this