Research Article
BibTex RIS Cite

Random Forest ve K-Nearest Neighbor Yöntemleri ile Günlük Akım Modellemesi

Year 2021, Volume: 14 Issue: 3, 914 - 925, 18.12.2021
https://doi.org/10.18185/erzifbed.949126

Abstract

Su; canlı yaşamı için vazgeçilemez bir doğal kaynaktır. Bu nedenle su kaynaklarının korunması ve kontrolü büyük önem arz etmektedir. Nehir akımı tahmini ve modellemesi; su kaynaklarının yönetimi, sulama faaliyetleri gibi durumlarda önem arz ettiği için sürekli araştırılmaya ve geliştirilmeye ihtiyaç duyulmuş bir konu olarak literatürde yer almaktadır. Tahmin ve modelleme için çok sayıda teknik kullanılmakta, yapılan çalışmaların gelişmesi, tekniklerin kıyaslanması ve eksik yönlerin görülmesi ile tahmin sonuçları giderek iyileşmektedir. Bu çalışmada da tahmin sonuçlarını değerlendirmek ve daha iyi olan tahmin yöntemini bulabilmek, yöntemlerin avantaj ve dezavantajlarını tespit edebilmek için makine öğrenmesi yöntemlerinden Random Forest ve K-Nearest Neighbors doğrusal olmayan regresyon modelleri kullanılmıştır. Ayrıca RF modeli için hiperparametre seçimi Random Search ve Grid Search yöntemleri kullanılarak da oluşturulup kıyaslaması yapılmıştır. Fırat havzasında yer alan iki istasyona ait 1981-2011 yılları için günlük akım verileri kullanılan çalışmada; rank analizi ile nihai sonuca ulaşılmış olup diğer modellere göre Random Forest modelinin hiperparametrelerinin belirlenebilmesi için Random Search uygulandığında daha iyi sonuç alındığı görülmüştür.

References

  • A.Kagoda, P., JohnNdiritu, CeliweNtuli, & BeasonMwaka. (2010). Application of radial basis function neural networks to short-term streamflow forecasting. Physics and Chemistry of the Earth, Parts A/B/C, 571-581. doi:https://doi.org/10.1016/j.pce.2010.07.021
  • A.M, A.-A., & S., S. (2016). Spatial mapping of artesian zone at Iraqi southern desert using a GIS-based random forest machine learning model. Modeling Earth Systems and Environment, 2(2), 96. https://doi.org/10.1007/s40808-016-0150-6
  • Altunkaynak, A., & Başakın, E. (2018). Zaman Serileri Kullanılarak Nehir Akım Tahmini ve Farklı Yöntemlerle Karşılaştırılması. Erzincan Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 11(1), 92-101. doi:10.18185/erzifbed.339781
  • Bergstra, J., & Bengio, Y. (2012). Random Search for Hyper-Parameter Optimization. Journal of Machine Learning Research, 13, 281-305.
  • Breiman, L. (2001). Random Forests. Machine Learning, 5-32.
  • DSI . (1981-2010). Akım Gözlem Yıllıkları : https://www.dsi.gov.tr/Sayfa/Detay/744
  • EIEI. (2000). Akım Gözlem Yıllığı. Ankara: T.C. Elektrik İşleri Etüt İdaresi .
  • G.H., H. (2014). An insight into extreme learning machines: random neurons, random features and kernels. Cognitive Computation, 6(3), 376-390. https://doi.org/10.1007/s12559-014-9255-2
  • Li, X., Sha, J., & Wang, Z. L. (2019). Comparison of daily streamflow forecasts using extreme learning machines and the random forest method. Hydrological Sciences Journal, 64(15), 1857-1866. https://doi.org/10.1080/02626667.2019.1680846
  • M.Chenga, F.Fanga, T.KinouchibI, M.Navonc, & C.C.Paina. (2020). Long lead-time daily and monthly streamflow forecasting using machine learning methods. Journal of Hydrology, 590, 125376. https://doi.org/10.1016/j.jhydrol.2020.125376
  • Modares, F., Araghinejad, S., & Ebrahimi, K. (2018). A Comparative Assessment of Artificial Neural Network, Generalized Regression Neural Network, Least-Square Support Vector Regression, and K-Nearest Neighbor Regression for Monthly Streamflow Forecasting in Linear and Nonlinear Conditions. Water Resources Management volume , 243-258.
  • Papacharalampous, G. A., & Tyralis, H. (2018). Evaluation of random forests and Prophet for daily streamflow forecasting. Advances in Geosciences, 201-208. https://doi.org/10.5194/adgeo-45-201-2018
  • Peterson, L. (2009). K-nearest neighbor. Scholarpedia, 1883. doi:10.4249/scholarpedia.1883
  • Tosunoglu, F., Hanay, Y. S., Cintas, E., & Özyer, B. (2020). Monthly Streamflow Forecasting Using Machine Learning. Erzincan Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 13(3), 1242-1251. doi:10.18185/erzifbed.780477
  • Were, K., TienBui, D., B.Dick, Ø., & RamSingh, B. (2015). A comparative assessment of support vector regression, artificial neural networks, and random forests for predicting and mapping soil organic carbon stocks across an Afromontane landscape. Ecological Indicators, 394-403.
  • Yaseen, Z. M., Sulaiman, S. O., Deo, R. C., & Chau, K.-W. (2019). An enhanced extreme learning machine model for river flow forecasting: State-of-the-art, practical application in water resource engineering area and future research direction. Journal of Hydrology, 120, 387-408.
  • Yenigün, K., & Gümüş, V. (2007). Fırat Havzası Akımlarında Görülen Trendlerin Nedenlerinin Araştırılması. V. Ulusal Hidroloji Kongresi, (s. 239-248). Ankara.
  • Z.M., Y., S.O., S., R.C., D., & K., C. (2019). An enhanced extreme learning machine model for river flow forecasting; State-of-art, practical application in water resources engineering area and future research direction. Journal of Hydrology, 120, 387-408.
  • Zhang, H., Zhou, J., Jahed Armaghani, D., Tahir, M., Pham, B., & Huynh, V. (2020). A Combination of Feature Selection and Random Forest Techniques to Solve a Problem Related to Blast-Induced Ground Vibration. Appl. Sci., 10 (3), 869. doi:https://doi.org/10.3390/app10030869

Daily Flow Modeling With Random Forest and K-Nearest Neighbor Methods

Year 2021, Volume: 14 Issue: 3, 914 - 925, 18.12.2021
https://doi.org/10.18185/erzifbed.949126

Abstract

Water is an indispensable natural resource for living life. Therefore, protection and control of water resources are of great importance. Since river flow estimation and modeling are very important in cases such as the management of water resources, irrigation, it is included in the literature as an issue that needs constant research and development. A large number of techniques are being used for estimation and modeling; thus, the estimation results are gradually improving with the development of the studies carried out, the comparison of techniques, and the determination and removal of the shortcomings. In this study, Random Forest and K-Nearest Neighbors nonlinear regression models, which are two of the machine learning methods, were used to evaluating the estimation results, to find the better estimation method, and to determine the advantages and disadvantages of these methods. In addition, Random Search and Grid Search methods were used to make the hyperparameter selection and comparison for the Random Forest model. In this study, in which daily flow data of 1981-2011 of the two stations in the Euphrates were used, and, when compared to other models, it was observed that better results were obtained when Random Search was applied to determine the hyperparameters of the Random Forest model.

References

  • A.Kagoda, P., JohnNdiritu, CeliweNtuli, & BeasonMwaka. (2010). Application of radial basis function neural networks to short-term streamflow forecasting. Physics and Chemistry of the Earth, Parts A/B/C, 571-581. doi:https://doi.org/10.1016/j.pce.2010.07.021
  • A.M, A.-A., & S., S. (2016). Spatial mapping of artesian zone at Iraqi southern desert using a GIS-based random forest machine learning model. Modeling Earth Systems and Environment, 2(2), 96. https://doi.org/10.1007/s40808-016-0150-6
  • Altunkaynak, A., & Başakın, E. (2018). Zaman Serileri Kullanılarak Nehir Akım Tahmini ve Farklı Yöntemlerle Karşılaştırılması. Erzincan Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 11(1), 92-101. doi:10.18185/erzifbed.339781
  • Bergstra, J., & Bengio, Y. (2012). Random Search for Hyper-Parameter Optimization. Journal of Machine Learning Research, 13, 281-305.
  • Breiman, L. (2001). Random Forests. Machine Learning, 5-32.
  • DSI . (1981-2010). Akım Gözlem Yıllıkları : https://www.dsi.gov.tr/Sayfa/Detay/744
  • EIEI. (2000). Akım Gözlem Yıllığı. Ankara: T.C. Elektrik İşleri Etüt İdaresi .
  • G.H., H. (2014). An insight into extreme learning machines: random neurons, random features and kernels. Cognitive Computation, 6(3), 376-390. https://doi.org/10.1007/s12559-014-9255-2
  • Li, X., Sha, J., & Wang, Z. L. (2019). Comparison of daily streamflow forecasts using extreme learning machines and the random forest method. Hydrological Sciences Journal, 64(15), 1857-1866. https://doi.org/10.1080/02626667.2019.1680846
  • M.Chenga, F.Fanga, T.KinouchibI, M.Navonc, & C.C.Paina. (2020). Long lead-time daily and monthly streamflow forecasting using machine learning methods. Journal of Hydrology, 590, 125376. https://doi.org/10.1016/j.jhydrol.2020.125376
  • Modares, F., Araghinejad, S., & Ebrahimi, K. (2018). A Comparative Assessment of Artificial Neural Network, Generalized Regression Neural Network, Least-Square Support Vector Regression, and K-Nearest Neighbor Regression for Monthly Streamflow Forecasting in Linear and Nonlinear Conditions. Water Resources Management volume , 243-258.
  • Papacharalampous, G. A., & Tyralis, H. (2018). Evaluation of random forests and Prophet for daily streamflow forecasting. Advances in Geosciences, 201-208. https://doi.org/10.5194/adgeo-45-201-2018
  • Peterson, L. (2009). K-nearest neighbor. Scholarpedia, 1883. doi:10.4249/scholarpedia.1883
  • Tosunoglu, F., Hanay, Y. S., Cintas, E., & Özyer, B. (2020). Monthly Streamflow Forecasting Using Machine Learning. Erzincan Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 13(3), 1242-1251. doi:10.18185/erzifbed.780477
  • Were, K., TienBui, D., B.Dick, Ø., & RamSingh, B. (2015). A comparative assessment of support vector regression, artificial neural networks, and random forests for predicting and mapping soil organic carbon stocks across an Afromontane landscape. Ecological Indicators, 394-403.
  • Yaseen, Z. M., Sulaiman, S. O., Deo, R. C., & Chau, K.-W. (2019). An enhanced extreme learning machine model for river flow forecasting: State-of-the-art, practical application in water resource engineering area and future research direction. Journal of Hydrology, 120, 387-408.
  • Yenigün, K., & Gümüş, V. (2007). Fırat Havzası Akımlarında Görülen Trendlerin Nedenlerinin Araştırılması. V. Ulusal Hidroloji Kongresi, (s. 239-248). Ankara.
  • Z.M., Y., S.O., S., R.C., D., & K., C. (2019). An enhanced extreme learning machine model for river flow forecasting; State-of-art, practical application in water resources engineering area and future research direction. Journal of Hydrology, 120, 387-408.
  • Zhang, H., Zhou, J., Jahed Armaghani, D., Tahir, M., Pham, B., & Huynh, V. (2020). A Combination of Feature Selection and Random Forest Techniques to Solve a Problem Related to Blast-Induced Ground Vibration. Appl. Sci., 10 (3), 869. doi:https://doi.org/10.3390/app10030869
There are 19 citations in total.

Details

Primary Language English
Subjects Engineering
Journal Section Makaleler
Authors

Hüseyin Yildirim Dalkılıç 0000-0002-4405-9341

Sefa Nur Yeşilyurt 0000-0001-6173-3038

Pijush Samui 0000-0003-2906-6479

Publication Date December 18, 2021
Published in Issue Year 2021 Volume: 14 Issue: 3

Cite

APA Dalkılıç, H. Y., Yeşilyurt, S. N., & Samui, P. (2021). Daily Flow Modeling With Random Forest and K-Nearest Neighbor Methods. Erzincan University Journal of Science and Technology, 14(3), 914-925. https://doi.org/10.18185/erzifbed.949126