Pattern Reduction Enhanced Ant Colony Optimization Clustering Algorithm

Hidayat, Dwi Taufik and Fatichah, Chastine and Ginardi, Raden Vinantius (2016) Pattern Reduction Enhanced Ant Colony Optimization Clustering Algorithm. PROSIDING ISEMANTIC 2016. pp. 317-322. ISSN 978-1-5090-2325-7

[img]
Preview
Text (ARTICLE)
ISEMANTIC 2016 - PATTERN REDUCTION ENHANCED ANT COLONY OPTIMIZATION CLUSTERING ALGORITHM.pdf

Download (1MB) | Preview
[img] Text (REVIEW / VALIDASI)
PEER REVIEW-PATTERN REDUCTION ENHANCED ANT COLONY OPTIMIZATION CLUSTERING ALGORITHM.pdf

Download (455kB)
[img] Text (SIMILARITY)
TURNITIN-PATTERN REDUCTION ENHANCED ANT COLONY OPTIMIZATION CLUSTERING ALGORITHM.pdf

Download (3MB)
Official URL: http://isemantic.dinus.ac.id/

Abstract

Ant Colony Optimization is the optimization method used for the analysis of the cluster. This method performs clustering data well. This method has limitations on computation time.This paper proposed a method of pattern reduction enhanced ant colony optimization to resolve the problem. Functions reduction pattern used to bypass loop system. This function ensures pheromone value if it meets certain iteration, so the next ant agent definitely choose the path that has been determined. The test is done through testing parameters and test accuracy. accuracy test compared with ant colony optimization method(ACO) with the results of the proposed method has the advantage of computing time. The test results show that the proposed method has 46 second faster than ACO method (73 second) for iris dataset, 141 second faster than ACO method (272 second) for wine dataset, and 150 second faster than ACO method (204 second) for synthetic dataset.

Item Type: Article
Uncontrolled Keywords: Pattern reduction enhanced ant colony algorithm; clustering; optimization
Subjects: A General Works > AI Indexes (General)
Divisions: PROSIDING & Call for Papers > Call For Papers
Depositing User: Risma Noviana
Date Deposited: 19 Mar 2018 03:37
Last Modified: 20 Nov 2019 03:56
URI: http://repository.widyakartika.ac.id/id/eprint/123

Actions (login required)

View Item View Item