Combining Artificial Algae AlgorithmtoArtificial Neural Networkfor Optimization of Weights

Yıl 2018, Cilt: 1 Sayı: 1, 37 - 44, 26.12.2018

Gülay Tezel , Sait Ali Uymaz , Esra Yel

Öz

Artificial Neural Network (ANN) is one of the most important artificial intelligent algorithms used for classification problems. The structure of ANN depends on the learning algorithm used for adjusting the weights between neurons of the layers according to the calculated error between model value and the real value. Recently the weights between layers in ANN has been optimized by using metaheuristic optimization algorithms. One of the recent high performance nonlinear optimization algorithms is Artificial Algae Algorithm (AAA) which is a bioinspired, successful, competitive and robust optimization algorithm. In this study, AAA was used as a tool for optimization of the weights in ANN algorithm. ANN and AAA was combined such that the training step of the ANN modeling to be performed by AAA. After training, ANN continues testing with the optimized weights. The established model combination (AAANN) was tested on three benchmarked datasets (Iris, Thyroid and Dermatology) of the UCI Machine Learning Repository to indicate the performance of this hybrid structure. The results were compared with MLP algorithm in terms of Mean Absolute Error (MAE). Accordingly, up to 96% reduction in mean MSE levels could be achieved by AAANN for all models.

Anahtar Kelimeler

Artificial Neural Networks, Artificial Algae Algorithm, backpropagation, classification algorithms, optimization

Kaynakça

• [1] L. Fausett, Fundamentals of Neural Networks Architectures Algorithms and Applications, Prentice Hall, Upper Saddle River, New Jersey, 1994.
• [2] S. Haykin, Neural Networks: A Comprehensive Foundation, New York: Macmillan, 1994.
• [3] V. G. Gudise ve G. K. Venayagamoorthy, "Comparison of Particle Swarm Optimization and Backpropagation as Training Algorithms for Neural Networks," Proc. of the 2003 IEEE Swarm Intelligence Symposium, 2003.
• [4] E. Momeni, R. Nazir, D. J. Armaghani ve H. Maizir, "Prediction of pile bearing capacity using a hybrid genetic algorithm-based ANN," Measurement, 2014, cilt 57, ss. 22–131.
• [5] D. J. Montana ve L. Davis, "Training Feedforward Neural Networks Using Genetic Algorithms," IEEE Int. Conf. on Systems, Man and Cybernetics, 2000.
• [6] B. A. Garro ve R. A. Vázquez, "Designing Artificial Neural Networks Using Particle Swarm Optimization Algorithms," Computational Intelligence and Neuroscience, 2015.
• [7] S. A. Uymaz, G. Tezel ve E. Yel, "Artificial algae algorithm (AAA) for nonlinear global optimization," Applied Soft Computing, cilt 31, ss. 153-171, 2015.
• [8] S. A. Uymaz, G. Tezel ve E. Yel, "Artificial algae algorithm with multi-light source for numerical optimization and applications," Biosystems, 2015, cilt 138, ss. 25-38.
• [9] X. Zhang, C. Wu, J. Li, X. Wang, Z. Yang, J. Lee ve K. Jung, "Binary artificial algae algorithm for multidimensional knapsack problems," Applied Soft Computing, 2016, cilt 43, ss. 583-595.
• [10] S. Korkmaz, A. Babalik ve M. S. Kiran, "An artificial algae algorithm for solving binary optimization problems," Int. J. of Machine Learning and Cybernetics, 2018, cilt 9(7), ss. 1233–1247.
• [11] M. Beskirli, I. Koc ve H. Kodaz, "Individual Parameter Selection Strategy for Artificial Algae Algorithm (AAA)," Journal of Computers, 2018, cilt 13(4), ss. 450-460.
• [12] R. A. Fisher, "The use of multiple measurements in taxonomic problems," Annual Eugenics, 1936, cilt 7(II), ss. 179-188.
• [13] J. R. Quinlan, P. J. Compton, K. A. Horn ve L. Lazarus, "Inductive knowledge acquisition: a case study," Proc. of the second Australian Conf. on the App. of Expert Systems, 1986, cilt 1, ss. 183-204.
• [14] H. A. Güvenir, G. Demiröz ve N. Ilter, "Learning differential diagnosis of erythemato-squamous diseases using voting feature intervals," Artificial Intelligence in Medicine, 1998, cilt 13, ss. 147.