Login | Register

GRUEL: An EL Reasoner Using General Purpose Computing on a Graphical Processing Unit

Title:

GRUEL: An EL Reasoner Using General Purpose Computing on a Graphical Processing Unit

Kharma, Myriam (2017) GRUEL: An EL Reasoner Using General Purpose Computing on a Graphical Processing Unit. Masters thesis, Concordia University.

[thumbnail of Kharma_MSc_S2017.pdf]
Preview
Text (application/pdf)
Kharma_MSc_S2017.pdf - Accepted Version
Available under License Spectrum Terms of Access.
3MB

Abstract

Abstract

GRUEL: AN EL REASONER USING GENERAL PURPOSE COMPUTING ON A GRAPHICAL PROCESSING UNIT

Myriam Kharma


The third installment of the worldwide web (Web 3.0) relies on an infrastructure of efficient reasoners, able to quickly infer new knowledge from existing data. We have created a reasoner using general purpose computing on a graphical processing unit (GPU) for the “Existential Language” description language, denoted with EL, that exploits the processing power of a GPU while attempting to solve a logic-based artificial intelligence problem.
To classify small ontologies, our system is nearly 70% faster than other reasoners we have compared it with. Currently far from perfect, GRUEL is planned to improve and expand, yet for the moment proves that better performance is achievable, and that general purpose computing on a GPU is a methodology worth exploring while developing the semantic web.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Computer Science and Software Engineering
Item Type:Thesis (Masters)
Authors:Kharma, Myriam
Institution:Concordia University
Degree Name:M. Comp. Sc.
Program:Computer Science
Date:April 2017
Thesis Supervisor(s):Haarslev, Volker
ID Code:982475
Deposited By: MYRIAM KHARMA
Deposited On:09 Jun 2017 15:00
Last Modified:18 Jan 2018 17:55
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Research related to the current document (at the CORE website)
- Research related to the current document (at the CORE website)
Back to top Back to top