Abstract
We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).
| Original language | English (US) |
|---|---|
| Pages (from-to) | 361-374 |
| Number of pages | 14 |
| Journal | Omega |
| Volume | 29 |
| Issue number | 4 |
| DOIs | |
| State | Published - Aug 1 2001 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Strategy and Management
- Management Science and Operations Research
- Information Systems and Management
Cite this
}
An empirical study of design and testing of hybrid evolutionary-neural approach for classification. / Pendharkar, Parag C.
In: Omega, Vol. 29, No. 4, 01.08.2001, p. 361-374.Research output: Contribution to journal › Article
TY - JOUR
T1 - An empirical study of design and testing of hybrid evolutionary-neural approach for classification
AU - Pendharkar, Parag C.
PY - 2001/8/1
Y1 - 2001/8/1
N2 - We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).
AB - We propose a hybrid evolutionary-neural approach for binary classification that incorporates a special training data over-fitting minimizing selection procedure for improving the prediction accuracy on holdout sample. Our approach integrates parallel global search capability of genetic algorithms (GAs) and local gradient-descent search of the back-propagation algorithm. Using a set of simulated and real life data sets, we illustrate that the proposed hybrid approach fares well, both in training and holdout samples, when compared to the traditional back-propagation artificial neural network (ANN) and a genetic algorithm-based artificial neural network (GA-ANN).
UR - http://www.scopus.com/inward/record.url?scp=0038232428&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0038232428&partnerID=8YFLogxK
U2 - 10.1016/S0305-0483(01)00031-7
DO - 10.1016/S0305-0483(01)00031-7
M3 - Article
AN - SCOPUS:0038232428
VL - 29
SP - 361
EP - 374
JO - Omega
JF - Omega
SN - 0305-0483
IS - 4
ER -