Research Article
BibTex RIS Cite

Zaman Serisi Tahmin Probleminin İmmün Plazma Programlama Kullanılarak Çözülmesi

Year 2021, , 219 - 224, 01.12.2021
https://doi.org/10.31590/ejosat.1010078

Abstract

Otomatik programlama, bir sistemde girdi ve çıktı değişkenleri arasındaki ilişkiyi model çıkararak açıklamaya çalışan bir makine öğrenmesi yaklaşımıdır. Zaman içerisinde sembolik regresyon, kümeleme, sınıflandırma, görüntü işleme, devre tasarımı, yol planlama, tahmin gibi birçok mühendislik problemlerini çözmeyi amaçlayan otomatik programlama yöntemleri geliştirilmiştir. Otomatik programlama yöntemlerinden birçoğu doğadan esinlenmektedir. Hızlı yayılan yeni Coronavirüs (COVID-19) salgınıyla mücadele edebilmek için farklı tedavi yöntemleri denenmektedir. İmmün plazma tedavisi, geçmişte birçok farklı salgında ve son olarak COVID-19’da etkili olduğu gösterilmiş tıbbi bir tedavi yöntemidir. İmmün plazma tedavi yaklaşımının uygulama aşamalarına dayanan İmmün Plazma Algoritması (Immune Plasma Algorithm, IPA) kısa süre önce önerilmiş bir meta-sezgisel algoritmadır. IPA, 2020 yılında tanıtılmış yeni bir algoritma olmasına rağmen farklı problemleri çözmek için çeşitli alanlarda uygulanmıştır. Bu çalışmada, IPA algoritmasını temel alan İmmün Plazma Programlama (Immune Plasma Programming, IPP) bir otomatik programlama yöntemi olarak tanıtılmıştır. IPP algoritmasının genel işleyişi IPA’nın aşamalarına benzerdir. Çözümlerin temsili ve iyileştirme mekanizması IPP’nin temel farklarıdır. IPA çözümleri sabit boyutlu diziler şeklinde ifade ederken, IPP çözümleri farklı derinliklere sahip olabilen parçalı ağaçlar olarak ifade eder. Ağaçların en küçük birimi düğümlerle temsil edilir. Düğümler, problemler için özel tanımlanan terminal kümesinden (x, y gibi değişkenler ve sabitler) ve fonksiyon kümesinden (aritmetik operatörler, mantıksal fonksiyonlar, matematiksel fonksiyonlar) seçilirler. Bu düğümlerin birleşimi ile çözümleri temsil eden ağaçlar oluşturulur. Çözümlerin iyileştirme mekanizması olarak Yapay Arı Koloni Programlama’da (Artificial Bee Colony Programming, ABCP) kullanılan bilgi paylaşım mekanizması IPP’ye uyarlanmıştır. Önerilen algoritmanın performansı, literatürde yaygın olarak kullanılan Box-Jenkins zaman serisi kullanılarak incelenmiştir. Çıkarılan modeller, en çok kullanılan otomatik programlama yöntemi ABCP ve Yapay Sinir Ağı modelleri ile kıyaslanmıştır. Sonuçlar, IPP’nin zaman serileri tahmin problemlerinde başarıyla kullanabileceğini göstermiştir.

Supporting Institution

-

Project Number

-

Thanks

Bu çalışmanın yürütülmesinde destek veren kıymetli meslektaşlarım Dr. Öğretim Üyesi Fırat İsmailoğlu’na ve Begüm Yetişkin’e saygılarımı sunar, teşekkür ederim.

References

  • Akdi, Y. (2003). Zaman Serileri Analizi.
  • Arslan, S., & Öztürk, C., (2019). Multi Hive Artificial Bee Colony Programming for high dimensional symbolic regression with feature selection. APPLIED SOFT COMPUTING, vol.78, 515-527.
  • Aslan, S., & Demirci, S. (2020). Immune Plasma Algorithm: A Novel Meta-Heuristic for Optimization Problems. IEEE Access, 8, 220227-220245. https://doi.org/10.1109/access.2020.3043174
  • Aslan, S., & Demirci, S. (2021). Performance Investigation of Parallel Immune Plasma Algorithm. Içinde 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA). 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA). IEEE. https://doi.org/10.1109/inista52262.2021.9548547
  • Biermann, A. W. (1985). Automatic programming: A tutorial on formal methodologies. Journal of Symbolic Computation, 1(2), 119-142. https://doi.org/10.1016/s0747-7171(85)80010-9
  • Box, G., Jenkins, G. (1970). Time Series Analysis: Forecasting and Control. Holden Day, San Francisco.
  • Cano, A., & Krawczyk, B. (2019). Evolving rule-based classifiers with genetic programming on GPUs for drifting data streams. Pattern Recognition, 87, 248-268. https://doi.org/10.1016/j.patcog.2018.10.024
  • Cascella M., Rajnik M., Cuomo A., Dulebohn, S. C., & Napoli R. D. (2020). Features, evaluation and treatment coronavirus (COVID-19). StatPearls [Internet], Stat Pearls Publishing.
  • Chen, Y., Yang, B., & Dong, J. (2004). Evolving Flexible Neural Networks Using Ant Programming and PSO Algorithm. Içinde Advances in Neural Networks – ISNN 2004 (ss. 211-216). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-28647-9_36
  • Golafshani, E. M. (2015). Introduction of Biogeography-Based Programming as a new algorithm for solving problems. Applied Mathematics and Computation, 270, 1-12. https://doi.org/10.1016/j.amc.2015.08.026
  • Görkemli, B., Öztürk, C., & Karaboğa, D. (2012). Yapay Arı Kolonisi Programlama ile Sistem Modelleme. Otomatik Kontrol Türk Milli Komitesi 2012 Ulusal Toplantısı (TOK2012). 857-860. Niğde, Turkey.
  • Huang, C., Wang, Y., Li, X., Ren, L., Zhao, J., Hu, Y., Zhang, L., Fan, G., Xu, J., Gu, X., Cheng, Z., Yu, T., Xia, J., Wei, Y., Wu, W., Xie, X., Yin, W., Li, H., Liu, M., … Cao, B. (2020). Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The Lancet, 395(10223), 497-506. https://doi.org/10.1016/s0140-6736(20)30183-5
  • Johnson, C. G. (2003). Artificial Immune System Programming for Symbolic Regression. Lecture Notes in Computer Science, 345-353. Springer Berlin Heidelberg. https://doi.org/10.1007/3-540-36599-0_32
  • Karaboğa, D., Öztürk, C., Karaboğa, N., & Görkemli, B. (2012). Artificial bee colony programming for symbolic regression, Information Sciences, 209, 1–15. Koza, J.R. (1992). Genetic Programming: On the Programming of Computers by Means of Natural Selection, Cambridge: MIT Press.
  • Li, Z. & Tanaka, G. (2021). Multi-Reservoir Echo State Networks with Sequence Resampling for Nonlinear Time-Series Prediction. Neurocomputing.
  • Roux, O., Fonlupt, C. (2000). Ant programming: or how to use ants for automatic programming, in 2nd International Workshop on Ant Algorithms (ANTS’2000), 121–129. Brussels, Belgium.
  • Sotto, L. F. D. P., de Melo, V. V., & Basgalupp, M. P. (2016). An improved λ-linear genetic programming evaluated in solving the Santa Fe ant trail problem. Proceedings of the 31st Annual ACM Symposium on Applied Computing. SAC 2016: Symposium on Applied Computing. https://doi.org/10.1145/2851613.2851669

Solving the Problem of Time Series Prediction Using Immune Plasma Programming

Year 2021, , 219 - 224, 01.12.2021
https://doi.org/10.31590/ejosat.1010078

Abstract

Automatic programming is a machine learning approach that attempts to explain the relationship between input and output variables in a system by extracting a model. Over time, automatic programming methods have been developed that aim to solve many engineering problems, such as symbolic regression, clustering, classification, image processing, circuit design, path planning, prediction. Most of the automatic programming methods are inspired by nature. To combat the rapidly spreading new coronavirus pandemic (COVID-19), various treatment methods are being tried. Immune plasma treatment is a medical treatment method that has proven to be effective in many different pandemics in the past and most recently in COVID-19. The Immune Plasma Algorithm (IPA) is a recently proposed meta-heuristic algorithm based on the implementation steps of the immune plasma treatment approach. Although IPA is a new algorithm, introduced in 2020, it has already been used in various fields to solve different problems. In this paper, the Immune Plasma Programming (IPP) automatic programming method based on IPA algorithm is presented for the first time. The general procedure of the IPP algorithm is similar to the stages of IPA. The representation of the solutions and the improvement mechanism are the main differences between the two algorithms. IPA expresses the solutions as fixed size arrays, while IPP represents the solutions as fragmented trees that can have different depths. The smallest unit of the trees is represented by nodes. The nodes are selected from a set of terminals (variables and constants such as x, y) and a set of functions (arithmetic operators, logical functions, mathematical functions) defined for the problems. Solution trees are created by combining these nodes. As the improvement mechanism of the solutions, the information sharing mechanism used in Artificial Bee Colony Programming (ABCP) was adapted to IPP. The performance of the algorithm was evaluated using the Box-Jenkins time series, which is widely used in the literature. The extracted models were compared with the most commonly used automatic programming method ABCP and Artificial Neural Network (ANN) models. The results showed that IPP can be successfully applied to time series prediction problems.

Project Number

-

References

  • Akdi, Y. (2003). Zaman Serileri Analizi.
  • Arslan, S., & Öztürk, C., (2019). Multi Hive Artificial Bee Colony Programming for high dimensional symbolic regression with feature selection. APPLIED SOFT COMPUTING, vol.78, 515-527.
  • Aslan, S., & Demirci, S. (2020). Immune Plasma Algorithm: A Novel Meta-Heuristic for Optimization Problems. IEEE Access, 8, 220227-220245. https://doi.org/10.1109/access.2020.3043174
  • Aslan, S., & Demirci, S. (2021). Performance Investigation of Parallel Immune Plasma Algorithm. Içinde 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA). 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA). IEEE. https://doi.org/10.1109/inista52262.2021.9548547
  • Biermann, A. W. (1985). Automatic programming: A tutorial on formal methodologies. Journal of Symbolic Computation, 1(2), 119-142. https://doi.org/10.1016/s0747-7171(85)80010-9
  • Box, G., Jenkins, G. (1970). Time Series Analysis: Forecasting and Control. Holden Day, San Francisco.
  • Cano, A., & Krawczyk, B. (2019). Evolving rule-based classifiers with genetic programming on GPUs for drifting data streams. Pattern Recognition, 87, 248-268. https://doi.org/10.1016/j.patcog.2018.10.024
  • Cascella M., Rajnik M., Cuomo A., Dulebohn, S. C., & Napoli R. D. (2020). Features, evaluation and treatment coronavirus (COVID-19). StatPearls [Internet], Stat Pearls Publishing.
  • Chen, Y., Yang, B., & Dong, J. (2004). Evolving Flexible Neural Networks Using Ant Programming and PSO Algorithm. Içinde Advances in Neural Networks – ISNN 2004 (ss. 211-216). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-28647-9_36
  • Golafshani, E. M. (2015). Introduction of Biogeography-Based Programming as a new algorithm for solving problems. Applied Mathematics and Computation, 270, 1-12. https://doi.org/10.1016/j.amc.2015.08.026
  • Görkemli, B., Öztürk, C., & Karaboğa, D. (2012). Yapay Arı Kolonisi Programlama ile Sistem Modelleme. Otomatik Kontrol Türk Milli Komitesi 2012 Ulusal Toplantısı (TOK2012). 857-860. Niğde, Turkey.
  • Huang, C., Wang, Y., Li, X., Ren, L., Zhao, J., Hu, Y., Zhang, L., Fan, G., Xu, J., Gu, X., Cheng, Z., Yu, T., Xia, J., Wei, Y., Wu, W., Xie, X., Yin, W., Li, H., Liu, M., … Cao, B. (2020). Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. The Lancet, 395(10223), 497-506. https://doi.org/10.1016/s0140-6736(20)30183-5
  • Johnson, C. G. (2003). Artificial Immune System Programming for Symbolic Regression. Lecture Notes in Computer Science, 345-353. Springer Berlin Heidelberg. https://doi.org/10.1007/3-540-36599-0_32
  • Karaboğa, D., Öztürk, C., Karaboğa, N., & Görkemli, B. (2012). Artificial bee colony programming for symbolic regression, Information Sciences, 209, 1–15. Koza, J.R. (1992). Genetic Programming: On the Programming of Computers by Means of Natural Selection, Cambridge: MIT Press.
  • Li, Z. & Tanaka, G. (2021). Multi-Reservoir Echo State Networks with Sequence Resampling for Nonlinear Time-Series Prediction. Neurocomputing.
  • Roux, O., Fonlupt, C. (2000). Ant programming: or how to use ants for automatic programming, in 2nd International Workshop on Ant Algorithms (ANTS’2000), 121–129. Brussels, Belgium.
  • Sotto, L. F. D. P., de Melo, V. V., & Basgalupp, M. P. (2016). An improved λ-linear genetic programming evaluated in solving the Santa Fe ant trail problem. Proceedings of the 31st Annual ACM Symposium on Applied Computing. SAC 2016: Symposium on Applied Computing. https://doi.org/10.1145/2851613.2851669
There are 17 citations in total.

Details

Primary Language Turkish
Subjects Engineering
Journal Section Articles
Authors

Sibel Arslan 0000-0003-3626-553X

Project Number -
Publication Date December 1, 2021
Published in Issue Year 2021

Cite

APA Arslan, S. (2021). Zaman Serisi Tahmin Probleminin İmmün Plazma Programlama Kullanılarak Çözülmesi. Avrupa Bilim Ve Teknoloji Dergisi(29), 219-224. https://doi.org/10.31590/ejosat.1010078