Feature selection for high-dimensional class-imbalanced data sets using Support Vector Machines

2014 ◽  
Vol 286 ◽  
pp. 228-246 ◽  
Author(s):  
Sebastián Maldonado ◽  
Richard Weber ◽  
Fazel Famili
Keyword(s):  
Feature Selection ◽  
Support Vector Machines ◽  
Imbalanced Data ◽  
High Dimensional ◽  
Support Vector ◽  
Data Sets ◽  
Imbalanced Data Sets ◽  
Vector Machines ◽  
Selection For

scholarly journals On relationships between imbalance and overlapping of datasets

10.29007/h71z ◽  
2020 ◽  
Author(s):  
Waleed Almutairi ◽  
Ryszard Janicki
Keyword(s):  
Support Vector Machines ◽  
Performance Indicators ◽  
Imbalanced Data ◽  
Nearest Neighbors ◽  
Support Vector ◽  
Classification Algorithms ◽  
Data Sets ◽  
K Nearest Neighbors ◽  
Imbalanced Data Sets ◽  
Vector Machines

The paper deals with problems that imbalanced and overlapping datasets often en- counter. Performance indicators as accuracy, precision and recall of imbalanced data sets, both with and without overlapping, are discussed and compared with the same performance indicators of balanced datasets with overlapping. Three popular classification algorithms, namely, Decision Tree, KNN (k-Nearest Neighbors) and SVM (Support Vector Machines) classifiers are analyzed and compared.

FEATURE SELECTION FOR SUPPORT VECTOR MACHINES USING GENETIC ALGORITHMS

2004 ◽  
Vol 13 (04) ◽  
pp. 791-800 ◽  
Author(s):  
HOLGER FRÖHLICH ◽  
OLIVIER CHAPELLE ◽  
BERNHARD SCHÖLKOPF
Keyword(s):  
Genetic Algorithms ◽  
Feature Selection ◽  
Support Vector Machines ◽  
Cross Validation ◽  
Support Vector ◽  
Generalization Error ◽  
New Approach ◽  
Vector Machines ◽  
Selection For ◽  
Natural Way

The problem of feature selection is a difficult combinatorial task in Machine Learning and of high practical relevance, e.g. in bioinformatics. Genetic Algorithms (GAs) offer a natural way to solve this problem. In this paper we present a special Genetic Algorithm, which especially takes into account the existing bounds on the generalization error for Support Vector Machines (SVMs). This new approach is compared to the traditional method of performing cross-validation and to other existing algorithms for feature selection.

Sign in / Sign up

Export Citation Format

Share Document