Abstract
Classification is inherently multi-objective problem. It is one of the most important tasks of data mining to extract important patterns from large volume of data. Traditionally, either only one objective is considered or the multiple objectives are accumulated to one objective function. In the last decade, Pareto-based multi-objective optimization approach have gained increasing popularity due to the use of multi-objective optimization using evolutionary algorithms and population-based search methods. Multi-objective optimization approaches are more powerful than traditional single-objective methods as it addresses various topics of data mining such as classification, clustering, feature selection, ensemble learning, etc. This paper proposes improved approach of non-dominated sorting algorithm II (NSGA II) for classification using neural network model by augmenting with local search. It tries to enhance two conflicting objectives of classifier: Accuracy and mean squared error. NSGA II is improved by augmenting backpropagation as a local search method to deal with the disadvantage of genetic algorithm, i.e. slow convergence to best solutions. By using backpropagation we are able to speed up the convergence. This approach is applied in various classification problems obtained from UCI repository. The neural network modes obtained shows high accuracy and low mean squared error.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
U. Maulik, A. Mukhopadhyay and S. Bandyopadhyay: Multiobjective Genetic Algorithms for Clustering—Applications in Data Mining and Bioinformatics. Springer, Berlin, Germany (2011).
Han and Kamber: Data Mining: Concepts and Techniques. San Francisco, CA, USA: Morgan Kaufmann (2006).
Kalyanmoy Deb: Multi-Objective Optimization Using Evolutionary Algorithms. ISBN 8126528044, 9788126528042, Wiley India (2010).
M. Butler and A. Daniyal: Multi-objective optimization with an evolutionary artificial neural network for financial forecasting. In: GECCO’09, 978-1-60558-325-9/09/07, pp. 1451-1457, ACM (2009).
K.O. Stanley and R. Mikkulainen: Evolving neural networks through augmenting topologies. Evolutionary Computation, 10(2) (2002).
H. Abbass: An evolutionary ANN approach for breast cancer diagnosis. In: Journal Artificial Intelligence in Medicine, vol. 25, Issue 3, pp. 265-281, Elsevier Science (2002).
Abbaas HA, Sarker R and Newton C: A Pareto-differential evolution approach for multi-objective optimization problems. In: IEEE congress on evolutionary computation, vol 2, pp. 971-978 (2001).
K. Deb, S. Agrawal, T. Meyarivan and A. Pratap: A fast and elitist multiobjective genetic algorithm: NSGA-II. In: IEEE Trans. Evol. Comput., vol. 6, no. 2, pp. 182–197, IEEE (2002).
A. Ibrahim, S. Shamsuddin, N. Ahmad and S. Qasem: Multi-objective genetic algorithm for training TTBPN. In: Proc. 4th Int. Conf. on Computing and Informatics (ICOCI 2013), pp. 32–38 (2013).
J. Ang, C. Goh, E. Teoh and A. Mamun: Multi-objective evolutionary recurrent neural networks for System Identification. pp. 1586–1592, IEEE (2007).
R. Furtuna, S. Curteanu and Leon: An elitist NSGA II algorithm enhanced with a neural network applied to the multi-objective optimization of a polysiloxane synthesis process. In: Journal Engineering Applications of Artificial Intelligence, vol. 24, Issue 5, pp. 772–785, Elsevier (2011).
M. Cruz-Ramirez, C. Hervas-Martinez, J. Fernandez, J. Briceno and M. Mata: Predicting patient survival after liver transplantation using evolutionary multi-objective artificial neural networks. In Journal Artificial Intelligence in Medicine, vol. 58, Issue 1, pp. 37–49, Elsevier (2013).
D. Gossard, B. Lartigue and F. Thellier: MOO of a building envelope for thermal performance using GAs and ANN. In: Journal of Energy and Buildings 67, pp. 253–260, Elsevier (2013).
C. Coello, D. Veldhuizen and G. Lamont: Evolutionary Algorithms for Solving Multi-Objective Problems. Gen. and Evol.Computation, 2nd ed., ISBN 0387367977, 9780387367972, Springer, Berlin/Heidelberg, Germany (2007).
K. Deb and N. Srinivas: Multiobjective optimization using nondominated sorting in genetic algorithms. In: Journal Evol. Comput., vol. 2, pp. 221–248 (1994).
L. Thiele and E. Zitzler: Multiobjective evolutionary algorithms: The strength Pareto approach. In: IEEE Trans. Evol. Comput., vol. 3, pp. 257–271 (1999).
E. Zitzler, L. Thiele and M. Laumanns: SPEA2: Improving the strength Pareto evolutionary algorithm. In: Proc. EUROGEN Int. Conf. on Evol. methods for Design, Opti. And control with appls. to industrial problems, pp. 95–100 (2001).
J. Knowles and D. Corne: The Pareto archived evolution strategy: A new baseline algorithm for Pareto MOO. In: Proc. IEEE Cong. Evol. Comput., pp. 98–105, IEEE (1999).
D. Corne, M. Oates and J. Knowles: The Pareto envelope based selection algorithm for multiobjective optimization. In: Proc. Conf. PPSN-VI, pp. 839–848 (2000).
D. Corne, J. Knowles, N. Jerram, and M. Oates: PESA-II: Region-based selection in evolutionary multiobjective optimization. In: Proc. GECCO, pp. 283–290 (2001).
University of California, Irvine (UCI) Repository of Machine Learning Databases https://archive.ics.uci.edu/ml/machine-learning-databases/.
C. Goh, E. Teoh, and K. C. Tan: Hybrid multi-objective evolutionary design for artificial neural networks. In: IEEE Trans. On neural networks, vol. 19, pp. 1531–1548, IEEE (2008).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer Science+Business Media Singapore
About this paper
Cite this paper
Mane, S., Sonawani, S., Sakhare, S. (2016). Hybrid Multi-objective Optimization Approach for Neural Network Classification Using Local Search. In: Saini, H., Sayal, R., Rawat, S. (eds) Innovations in Computer Science and Engineering. Advances in Intelligent Systems and Computing, vol 413. Springer, Singapore. https://doi.org/10.1007/978-981-10-0419-3_21
Download citation
DOI: https://doi.org/10.1007/978-981-10-0419-3_21
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-0417-9
Online ISBN: 978-981-10-0419-3
eBook Packages: EngineeringEngineering (R0)