Scribd
Upload
40 views2 pages

Study of Intelligent Evolutionary Algorithms For Solving High Dimensional Problems

The document proposes a multi-objective approach for feature selection to improve clustering and classification of high dimensional data. It applies this approach to Growing Hierarchical Self-Organizing Maps (GHSOMs) and evaluates it on the DARPA/NSL-KDD datasets. The main methods used are feature selection as a multi-objective optimization problem and a multi-objective genetic algorithm to identify multiple Pareto solutions for feature selection. The goal is to not only improve classification accuracy but also the model's generalization ability.

Uploaded by

Ashok Prajya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
40 views2 pages

Study of Intelligent Evolutionary Algorithms For Solving High Dimensional Problems

The document proposes a multi-objective approach for feature selection to improve clustering and classification of high dimensional data. It applies this approach to Growing Hierarchical Self-Organizing Maps (GHSOMs) and evaluates it on the DARPA/NSL-KDD datasets. The main methods used are feature selection as a multi-objective optimization problem and a multi-objective genetic algorithm to identify multiple Pareto solutions for feature selection. The goal is to not only improve classification accuracy but also the model's generalization ability.

Uploaded by

Ashok Prajya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Project Title:

Study of Intelligent Evolutionary Algorithms


For Solving High Dimensional Problems
Abstract:
Feature selection is the key for all the Clustering and classification problems and
High Dimensional Algorithms by choosing an adequate feature subset, a dataset
dimensionality reduction is allowed, thus it decrease the classification
computational complexity, and improves the classifier performance and optimize
by avoiding redundant or irrelevant features. Feature selection can be formally
defined as an optimization problem with only one objective, that is, the
classification accuracy obtained by using the selected feature subset. In Multi-
objective approach the Feature Selection either select features that not only
improve the classification accuracy, but also the generalization capability in case
of supervised classifiers, or counterbalance the bias toward lower or higher
numbers of features that present some methods used to validate the
clustering/classification in case of unsupervised classifiers.

The main contribution of this paper is a multi-objective approach for feature


selection and its application to an unsupervised clustering procedure based on
Growing Hierarchical Self-Organizing Maps (GHSOMs) that includes a new method
for unit labeling and efficient determination of the winning unit. This multi-
objective approach makes it possible not only to differentiate between normal
and anomalous traffic but also among different anomalies. The efficiency of our
proposals has been evaluated by using the well-known DARPA/NSL-KDD datasets
that contain extracted features and labeled attacks from around 2 million
connections.

Methods Used:
1. Feature selection:-
Feature selection can be considered as a multi-objective problem with
conflicting goals. In credit scoring, these goals are: increasing the model
profitability, reducing the data acquisition costs and improving the
interpretability of the model.

2. Multi-Objective Optimization:-
Multi-objective optimization considers optimization problems involving more
than one objective function to be optimized simultaneously. It is useful when
optimal decisions need to be taken in the presence of trade-offs between two
or more conflicting objectives.

Algorithm used:
Multi-Objective Genetic Algorithm:-
Multi-objective genetic algorithm (MOGA) is a direct search method for
multi-objective optimization problems. It is based on the process of the
genetic algorithm; the population-based property of the genetic algorithm
is well applied in MOGAs. Comparing with the traditional multi-objective
algorithm whose aim is to find a single Pareto solution, the MOGA intends
to identify numbers of Pareto solutions.

Formulae used:
Let x be a vector containing p decision variables.

The optimization problem with n objective functions is expressed as:

Minimize/maximize fi(x) for i=1, 2, 3, ........,n

subject to gj(x)<0 ,j=1,2,3....J, gk(x)=0,k=1,2,3.....K.

You might also like