SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training

Abstract: Artificial neural networks (ANNs) are computational artificial intelligence techniques inspired from the biological nervous system. They mimic the human brain's abilities like thinking, inferring and making decisions. While doing so, they benefit from data, knowledge and experience, just like the human brain. ANNs are very successful in learning and predicting. They learn by training the examples and do predictions by using the learned knowledge. Therefore network's training/learning process is very important in ANNs and affects the success of the network. Many methods have been developed so far in the literature for training ANNs and among these methods meta-heuristic algorithms provide very successful results. In this study, social spider (SS) and ion motion (IM) algorithms are hybridized and used in the training of ANNs to take advantage of the strengths of the algorithms. SS algorithm is inspired from the behaviors of social spiders in nature and is a good candidate to find optimum points. IM algorithm is inspired from the motion of ions and avoids from trapping into local optima. In the study, the crystal phase of the IM algorithm is embedded to the SS algorithm in order to increase the search ability of SS algorithm and to avoid trapping into local optima. ANNs are trained by the hybrid SSIM algorithm on 3 classification benchmark problems and compared with IM and SS algorithms. The obtained results show that the hybrid SSIM algorithm has potential on training ANNs.

Keywords: Social Spider Algorithm, Ions Motion Algorithm, Artificial Neural Network, Classification

Publication details: Conference abstract presented at the 4th International Conference on Engineering and Natural Sciences (ICENS 2018), 2-6 May 2018, Kyiv, Ukraine. Published in: Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018), pp. 221. Publisher: ICENS. ISBN: 978-605-67955-2-7.

How to cite / Kaynak gösterimi

APA 7: Gülmez, B., & Kulluk, S. (2018, May). SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training [Conference abstract]. In Ö. Çınar (Ed.), Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018) (p. 221). ICENS. 4th International Conference on Engineering and Natural Sciences (ICENS 2018), Kyiv, Ukraine. https://doi.org/10.5281/zenodo.23036882

IEEE: B. Gülmez and S. Kulluk, "SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training," in Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018), Kyiv, Ukraine, May 2018, p. 221, ISBN: 978-605-67955-2-7, doi: 10.5281/zenodo.23036882.

Chicago: Gülmez, Burak, and Sinem Kulluk. 2018. "SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training." In Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018), 221. Kyiv, Ukraine: ICENS. https://doi.org/10.5281/zenodo.23036882.

MLA 9: Gülmez, Burak, and Sinem Kulluk. "SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training." Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018), ICENS, 2018, p. 221. https://doi.org/10.5281/zenodo.23036882.

Harvard: Gülmez, B. and Kulluk, S. (2018) 'SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training', in Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018). Kyiv, Ukraine: ICENS, pp. 221. doi: 10.5281/zenodo.23036882.

BibTeX:

@inproceedings{Gulmez2018ssim,
  author    = {Gülmez, Burak and Kulluk, Sinem},
  title     = {SSIM: A Hybrid of Social Spider and Ions Motion Algorithms and Its Application to ANN Training},
  booktitle = {Book of Abstracts of the IV International Conference on Engineering and Natural Sciences (ICENS 2018)},
  year      = {2018},
  month     = {may},
  pages     = {221},
  address   = {Kyiv, Ukraine},
  publisher = {ICENS},
  isbn      = {978-605-67955-2-7},
  doi       = {10.5281/zenodo.23036882},
  url       = {https://doi.org/10.5281/zenodo.23036882}
}