BibTeX
@MISC{Sameh_datamining,
author = {Ahmed Sameh and Khalid Magdy},
title = {Data Mining Ant Colony for Classifiers},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract – Self-organizing Ant Colony Optimization (ACO) is a technique that is inspired by the behavior of the ants as social insect that work together to accomplish a common goal using wisdom of the crowd. ACO is one of the algorithms that put swarm intelligence into action. Swarm intelligence, which is based on the idea of collective behavior, has occupied ACO in various fields and problem solving domains. Data mining is one of the domains where ACO has been applied successfully and provided scalable solutions. In this paper, we describe a knowledge discovery classification technique based on ACO. AntMiner, first proposed in [5], is a rule induction algorithm that occupies collective intelligence to construct classification rules. Experimental results are shown as the AntMiner+ is implemented with different variations inspired from discrete optimization, fuzzy rule induction, self-organizing map (SOM), dimensionality reduction, parallel simultaneous rule learning and tested on different datasets. Moreover, further combinations of these variations that produced enhancement are also proposed and tested.
Keyphrases
data mining ant colony swarm intelligence rule induction collective intelligence scalable solution data mining different variation dimensionality reduction self-organizing map simultaneous rule learning common goal various field social insect knowledge discovery classification technique discrete optimization classification rule different datasets abstract self-organizing ant colony optimization problem solving domain collective behavior experimental result fuzzy rule induction
