Customizing FP-growth algorithm to parallel mining with Charm++ library

Marek Puścian

doi:10.1117/12.2281037

7 August 2017 Customizing FP-growth algorithm to parallel mining with Charm++ library

Marek Puścian

Proceedings Volume 10445, Photonics Applications in Astronomy, Communications, Industry, and High Energy Physics Experiments 2017; 104452L (2017) https://doi.org/10.1117/12.2281037
Event: Photonics Applications in Astronomy, Communications, Industry, and High-Energy Physics Experiments 2017, 2017, Wilga, Poland

Abstract

This paper presents a frequent item mining algorithm that was customized to handle growing data repositories. The proposed solution applies Master Slave scheme to frequent pattern growth technique. Efficient utilization of available computation units is achieved by dynamic reallocation of tasks. Conditional frequent trees are assigned to parallel workers basing on their workload. Proposed enhancements have been successfully implemented using Charm++ library. This paper discusses results of the performance of parallelized FP-growth algorithm against different datasets. The approach has been illustrated with many experiments and measurements performed using multiprocessor and multithreaded computer.

Citation Download Citation

Marek Puścian "Customizing FP-growth algorithm to parallel mining with Charm++ library", Proc. SPIE 10445, Photonics Applications in Astronomy, Communications, Industry, and High Energy Physics Experiments 2017, 104452L (7 August 2017); https://doi.org/10.1117/12.2281037

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available