GPU-parallelized artificial bee colony algorithm (GP-ABC) in data clustering application

GPU-parallelized artificial bee colony algorithm (GP-ABC) in data clustering application

Artificial Bee Colony (ABC) optimization and k-means algorithm are popularly used in data clustering application due to their accuracy and simplicity. However, as the number of dimension and data increases, program complexity may increase much further and ABC will execute in much slower time. This p...

Full description

Saved in:
Bibliographic Details
Main Author: Anggacipta, Gerry
Other Authors: Kyle Rupnow
Format: Final Year Project
Language:English
Published: 2014
Subjects:
DRNTU::Engineering::Computer science and engineering::Computing methodologies::Pattern recognition
Online Access:http://hdl.handle.net/10356/59878
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Artificial Bee Colony (ABC) optimization and k-means algorithm are popularly used in data clustering application due to their accuracy and simplicity. However, as the number of dimension and data increases, program complexity may increase much further and ABC will execute in much slower time. This project proposes a novel parallelization model on ABC called ‘GPU-parallelized Artificial Bee Colony (GP-ABC)’ algorithm in order to achieve speedup relatively to its normal sequential program execution. Testing has been done on several datasets from UCI Machine Learning repository such as Iris and Wine datasets. The results were encouraging and outperformed the ordinary ABC algorithm in terms of processing time.

Similar Items