New Rough Set Attribute Reduction Algorithm based on Grey Wolf Optimization
-
Upload
aboul-ella-hassanien -
Category
Engineering
-
view
470 -
download
2
Transcript of New Rough Set Attribute Reduction Algorithm based on Grey Wolf Optimization
New Rough Set Attribute
Reduction Algorithm based on Grey Wolf Optimization
Waleed Yamany* and Aboul Ella Hassanien
*Faculty of Computers and Information, Fayoum University and SRGE Member
Egyptsceince.net Bio-inspiring and evolutionary computation: Trends, applications and open issues workshop, 7 Nov. 2015 Faculty of
Computers and Information, Cairo University
Agenda Introduction Rough Set Grey Wolf Optimization (GWO) The Proposed System of Rough Set
and GWO Experimental Results Conclusions & Future Work
Introduction Feature selection is one of the most essential problems in
the fields of data mining, machine learning and pattern
recognition.
The main purpose of feature selection is to determine a
minimal feature subset from a problem domain while
retaining a suitably high accuracy in representing the
original features.
Rough Set Rough set theory can transact with uncertainty and vagueness in
data analysis. It has been widely applied in many fields such as
data mining, machine learning,.
Rough set theory provides a mathematical tool to find out data
dependencies and reduce the number of features included in
dataset by purely structural method.
It is a formal approximation of a crisp set in terms of a pair of
sets which give the lower and the upper approximation of the
original set
Grey Wolf Optimization (GWO)
Grey wolf optimizer (GWO) is a population based meta-heuristics algorithm simulates the leadership hierarchy and hunting mechanism of gray wolves in nature .
We consider the fittest solution as the alpha , and the second and the third fittest solutions are named beta and delta , respectively.
In the mathematical model of hunting behavior of grey wolves, we assumed the alpha , beta and delta have better knowledge about the potential location of prey.
Fitness Function
We implement the BA-RSFS feature selection
algorithms in MatLab 7.8. The computer used to
get results is Intel (R), 2.1 GHz CPU; 2 MB
RAM and the system is Windows 7 Professional.
The dataset used for experiments were
downloaded from UCI- Machine Learning
Repository.
Experimental Results:Specifications of Used Computer
Experimental Results:
Experimental Results:
Conclusions The goal of this paper was to propose a hybrid GWO with Rough set
feature selection method to select a smaller number of features and
achieving similar or even better classification performance than
using all features.
GWO proves performance advance in both classification accuracy
and feature reduction over common methods such as PSO and GA.
For further questions:
Waleed [email protected]