Classification of IRIS Data Set with Arithmetic Optimization Algorithm and Statistical Results
Öz
The aim of this paper is to optimize the classification performance with the arithmetic optimization algorithm (AOA), one of the swarm-based intelligent algorithms, by using the multilayer perceptron (MLP) model, which is an artificial neural network architecture. Model training is provided by IRIS flower data, which is widely used. AOA is a metaheuristic optimization method inspired by basic arithmetic functions consisting of discovery and exploitation phases. The MLP model is structured to consist of input, hidden, and output layers and is trained to classify the types of flowers in the IRIS dataset. The model"s performance was evaluated using statistical metrics such as accuracy, recall, and F1 score. Simulations were carried out using the MATLAB package program. When the results were examined, the average accuracy rate of the model was measured as 96.7%. The recall rate was 96.0% and the F1 score was 96.3%. These results show that hybridizing heuristic algorithms to AI network models can produce effective and efficient results in complex datasets.
Anahtar Kelimeler
Arithmetic Optimization Algorithm, Artificial Neural Networks, Multilayer Perceptrons, Optimization
Kaynakça
- [1] D. Izci, S. Ekinci, E. Eker, and M. Kayri, “Improved Manta Ray foraging optimization using opposition-based learning for optimization problems,” 2022 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA), pp. 1–6, Jun. 2020, doi: 10.1109/hora49412.2020.9152925.
- [2] P. Bansal, S. Gupta, S. Kumar, S. Sharma, and S. Sharma, “MLP-LOA: A metaheuristic approach to design an optimal multilayer perceptron,” Soft Computing, vol. 23, no. 23, pp. 12331–12345, Jan. 2019, doi: 10.1007/s00500-019-03773-2.
- [3] R. Yu, W. Yu, and X. Wang, “KAN or MLP: A fairer comparison,” arXiv preprint, arXiv:2407.16674, Aug. 17, 2024, doi: 10.48550/arXiv.2407.16674.
- [4] T. Hoya, “Review of the two existing artificial neural network models—Multilayer perceptron and probabilistic neural networks,” in Syntactic Networks—Kernel Memory Approach, vol. 1157, Studies in Computational Intelligence, Cham, Switzerland: Springer, 2024, pp. 17–34, doi: 10.1007/978-3-031-57312-5_2.
- [5] P. Bansal, S. Kumar, S. Pasrija, and S. Sharma, “A hybrid grasshopper and new cat swarm optimization algorithm for feature selection and optimization of multi-layer perceptron,” Soft Computing, vol. 24, pp. 15463–15489, Oct. 2020, doi: 10.1007/s00500-020-04877-w.
- [6] M. Gerşil and T. Palamutçuoğlu, “Ders çizelgeleme probleminin melez genetik algoritmalar ile performans analizi,” Niğde Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, vol. 6, no. 1, pp. 242–262, Apr. 2013.
- [7] E. Küçüksille and M. Tokmak, “Yapay Arı Kolonisi Algoritması Kullanarak Otomatik Ders Çizelgeleme,” Süleyman Demirel University Journal of Natural and Applied Sciences, vol. 15, no. 3, pp. 203–210, Jul. 2011.
- [8] F. A. Şenel, A. S. Yüksel, F. Gökçe, and T. Yiğit, “Gri kurt optimizasyon algoritması ile iki boyutlu dizilim yazılımının geliştirilmesi,” Journal of Balıkesir University Institute of Science and Technology, vol. 20, no. 2, pp. 293–306, Dec. 2018, doi: 10.25092/baunfbed.433321.
- [9] S. Ekinci, B. Hekimoğlu, A. Demirören, and S. Kaya, “Harris Hawks Optimization Approach for Tuning of FOPID Controller in DC-DC Buck Converter,” in Proc. 2019 International Artificial Intelligence and Data Processing Symposium (IDAP), Malatya, Turkey, 2019, pp. 1–9, doi: 10.1109/IDAP.2019.8875992.
- [10] S. Ekinci, B. Hekimoğlu, E. Eker, and D. Sevim, “Hybrid Firefly and Particle Swarm Optimization Algorithm for PID Controller Design of Buck Converter,” in Proc. 2019 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), Ankara, Turkey, 2019, pp. 1–6, doi: 10.1109/ISMSIT.2019.8932733.