Havzaların Benzerliklerini Tanımlamada Alternatif Bir Yaklaşım: Hiyerarşik Kümeleme Yöntemi Uygulaması

Makine öğrenmesi yöntemleri günümüzde birçok alanda kullanımını yaygınlaştırmış ve yerini sağlamlaştırmıştır. Denetimli, denetimsiz ve takviyeli öğrenme olmak üzere üç ana kola ayrılan makine öğrenmesi süreçleri, araştırmacıların gözle fark edemediği bağlantıları bulmada veya uzun süreli hesaplama gerektiren durumlarda ön plana çıkmaktadırlar. Denetimsiz öğrenme yöntemleri, etiketlerinin bulunmadığı verilerdeki kalıpları veya yapıyı keşfetmek için kullanılan makine öğrenmesi yöntemleridir. Hiyerarşik kümeleme süreci en önde gelen denetimsiz öğrenme yöntemlerinden birisidir. Bu çalışma havzaların benzerliklerini tanımlama da kullandığımız sürece alternatif bir yöntem sunmak amacıyla yürütülmüştür. Önerilen yöntemin avantajları arasında veri setinde yer alan tüm havzaların birbirleri ile olan ilişkilerini ortaya koyması, veri setindeki gürültüye daha az duyarlı olması, az havza içeren uygulamalarda daha kullanışlı olması ve küme içi tutarlılığı sağlamada araştırmacıya esneklik tanımasıdır. Çalışmada Türkiye’nin kuzeyinde bulunan bazı havzaların hidrolojik müdahale birimleri (HRU) görüntüleri ve hiyerarşik kümeleme yaklaşımı kullanılarak kümelenmesi incelenmiştir ve birbirine en çok benzeyen iki havzanın Ereğli ve Çaykıyı havzaları olduğu anlaşılmıştır. Havzaların birbirleri ile olan ilişkilerini ortaya çıkarmak için mesafe matrisi hazırlanmıştır. Ayrıca bağımsız kümeler oluşturmak için dendrogramın kesme mesafesi seçiminde dört farklı istatistiksel yaklaşım kullanılmıştır. İstatistiki yöntemlerin önerdiği küme sayıları içerisine kalmak şartı ve küme içi homojenliği korumak amacıyla 6 ayrı küme oluşturulmuş ve havzaların kümelere bağlı dağılımı gösterilmiştir. Bu çalışma havzaların HRU görüntülerine göre hidrolojik benzerliklerine dayanarak kümelenmelerinde alternatif bir bakış açısı sunmaktadır.

An Alternative Approach in Defining the Similarity of Catchments: Application of Hierarchical Clustering Method

Machine learning methods have widespread their use and strengthened their places in many areas. These procedures can be divided into three main branches as supervised, unsupervised, and reinforced learning and assist researchers to find connections in the data that cannot be seen or in situations that require long-term computation. Unsupervised methods are used to discover patterns or structures in data which does not have any labels. The hierarchical clustering process is one of the leading unsupervised methods. This study is an alternative approach to the process we used in defining the similarities of basins. The advantages of the proposed method are; reveals the relations between all basins, is less sensitive to noise in the data set, is more useful in applications with fewer basins, and is flexible in ensuring intra-cluster consistency. Some basins located in the north of Turkey based on their hydrological response units (HRU) images were hierarchically clustered and found out that and the two basins most similar to each other are the Ereğli and Çaykıyı catchments. A distance matrix was prepared to reveal the relations of the basins. Besides, to create independent clusters, four different statistical approaches were used to select the cut-off height of the dendrogram. To stay within the cluster numbers suggested by statistical methods and to ensure cluster homogeneity, 6 separate clusters were created. The distribution of the catchments depending on the clusters was illustrated. This study provides an alternative perspective for the clustering of basins based on HRU images.

PDF

___

Abdalla, A., Cen, H. Y., El-manawy, A. and He, Y. 2019. Infield oilseed rape images segmentation via improved unsupervised learning models combined with supreme color features. Computers and Electronics in Agriculture, 162, 1057-1068.
Asres, M. T. and Awulachew, S. B. 2010. SWAT based runoff and sediment yield modelling: a case study of the Gumera watershed in the Blue Nile basin. Ecohydrology & Hydrobiology, 10(2), 191-199.
Aytaç, E. 2020. Unsupervised learning approach in defining the similarity of catchments: Hydrological response unit based k-means clustering, a demonstration on Western Black Sea Region of Turkey. International Soil and Water Conservation Research, 8(3), 321-331.
Carcillo, F., Le Borgne, Y.-A., Caelen, O., Kessaci, Y., Oblé, F. and Bontempi, G. 2019. Combining unsupervised and supervised learning in credit card fraud detection. Information Sciences, 557, 317-331.
Castellanos-Garzón, J. A., Costa, E., Jaimes S, J. L. and Corchado, J. M. 2019. An evolutionary framework for machine learning applied to medical data. Knowledge-Based Systems, 185, 1-8.
Castro, L., Wasserman, E. A. and Lauffer, M. 2018. Unsupervised learning of complex associations in an animal model. Cognition, 173, 28-33.
Chen, J., Yousefi, M., Zhao, Y., Zhang, C. P., Zhang, S., Mao, Z. X., Peng, M. S. and Han, R. P. 2019. Modelling ore-forming processes through a cosine similarity measure: Improved targeting of porphyry copper deposits in the Manzhouli belt, China. Ore Geology Reviews, 107, 108-118.
Chen, J., Zeng, Y., Li, Y. and Huang, G.-B. 2019. Unsupervised feature selection based extreme learning machine for clustering. Neurocomputing, 386, 198-207.
Crosta, A. P., Sabine, C. and Taranik, J. V. 1998. Hydrothermal alteration mapping at Bodie, California, using AVIRIS hyperspectral data. Remote Sensing of Environment, 65(3), 309-319.
de Aguiar Neto, Fernando S., da Costa, Arthur F., Manzato, Marcelo G., Campello, Ricardo J. G. B. 2020. Pre-processing approaches for collaborative filtering based on hierarchical clustering. Information Sciences, 532, 172-191.
Dumont, M., Reninger, P. A., Pryet, A., Martelet, G., Aunay, B. and Join, J. L. 2018. Agglomerative hierarchical clustering of airborne electromagnetic data for multi-scale geological studies. Journal of Applied Geophysics, 157, 1-9.
Ferreira, A. D., Freitas, D. M., da Silva, G. G., Pistori, H. and Folhes, M. T. 2019. Unsupervised deep learning and semi-automatic data labeling in weed discrimination. Computers and Electronics in Agriculture, 165, 1-11.
Godec, P., Pancur, M., Ilenic, N., Copar, A., Strazar, M., Erjavec, A., Pretnar, A., Demsar, J., Staric, A., Toplak, M., Zagar, L., Hartman, J., Wang, H., Bellazzi, R., Petrovic, U., Garagna, S., Zuccotti, M., Park, D., Shaulsky, G. and Zupan, B. 2019. Democratized image analytics by visual programming through integration of deep models and small-scale machine learning. Nature Communications, 10, 1-7.
Govender, P. and Sivakumar, V. 2020. Application of k-means and hierarchical clustering techniques for analysis of air pollution: A review (1980-2019). Atmospheric Pollution Research, 11(1), 40-56.
Gungor, O. and Goncu, S. 2013. Application of the soil and water assessment tool model on the Lower Porsuk Stream Watershed. Hydrological Processes, 27(3), 453-466.
Haeberle, H. S., Helm, J. M., Navarro, S. M., Karnuta, J. M., Schaffer, J. L., Callaghan, J. J., Mont, M. A., Kamath, A. F., Krebs, V. E. and Ramkumar, P. N. 2019. Artificial Intelligence and Machine Learning in Lower Extremity Arthroplasty: A Review. Journal of Arthroplasty, 34(10), 2201-2203.
Han, D. H., Lee, S. and Seo, D. C. 2020. Using machine learning to predict opioid misuse among US adolescents. Preventive Medicine, 130, 1-7.
Haynes, P. 2012. Case based methods and national comparisons. Public Policy Beyond the Financial Crisis: An International Comparative Study. London, Routledge, 120-125.
Huang, F. R., Zhang, X. M., Li, Z. J., Zhao, Z. H. and He, Y. Y. 2018. From content to links: Social image embedding with deep multimodal model. Knowledge-Based Systems, 160, 251-264.
Jafarzadegan, M., Safi-Esfahani, F. and Beheshti, Z. 2019. Combining hierarchical clustering approaches using the PCA method. Expert Systems with Applications, 137, 1-10.
Janert, P. K. 2010. Data Analysis with Open Source Tools, O’Reilly Media, Inc, Cambridge, 319-324.
Janez Demšar, B. Z. 2013. Orange: Data Mining Fruitful and Fun - A Historical Perspective. Informatica, 37, 55–60.
Kaushik, M. and Mathur, B. 2014. Comparative Study of K-Means and Hierarchical Clustering Techniques. International jJournal of Software & Hardware Research in Engineering, 2(6), 93-98.
Kotsiantis, S. B. 2007. Supervised Machine Learning: A Review of Classification Techniques. Emerging Artificial Intelligence Applications in Computer Engineering, 160, 3-24.
Li, K. H., Ma, Z. J., Robinson, D. and Ma, J. 2018. Identification of typical building daily electricity usage profiles using Gaussian mixture model-based clustering and hierarchical clustering. Applied Energy, 231, 331-342.
Loreti, D., Lippi, M. and Torroni, P. 2020. Parallelizing Machine Learning as a service for the end-user. Future Generation Computer Systems, 105, 275-286.
Manning, C. D. and Schütze, H. 1999. Foundations of statistical natural language processing, MIT Press, Cambridge, 417-452.
Naik, A. and Samant, L. 2016. Correlation review of classification algorithm using data mining tool: WEKA, Rapidminer, Tanagra, Orange and Knime. International Conference on Computational Modelling and Security (Cms 2016), 85, 662-668.
Perret, B., Cousty, J., Guimaraes, S. J. F., Kenmochi, Y. and Najman, L. 2019. Removing non-significant regions in hierarchical clustering and segmentation. Pattern Recognition Letters, 128, 433-439.
Ramkumar, P. N., Haeberle, H. S., Bloomfield, M. R., Schaffer, J. L., Kamath, A. F., Patterson, B. M. and Krebs, V. E. 2019. Artificial Intelligence and Arthroplasty at a Single Institution: Real-World Applications of Machine Learning to Big Data, Value-Based Care, Mobile Health, and Remote Patient Monitoring. The Journal of Arthroplasty, 34(10), 2204-2209.
Reddy, M. V., Vivekananda, M. and Satish, R. U. V. N. 2017. Divisive Hierarchical Clustering with K-means and Agglomerative Hierarchical Clustering. International Journal of Computer Science Trends and Technology, 5(5), 6-11.
Rocha, J., Roebeling, P. and Rial-Rivas, M. E. 2015. Assessing the impacts of sustainable agricultural practices for water quality improvements in the Vouga catchment (Portugal) using the SWAT model. Science of the Total Environment, 536, 48-58.
Sahu, S., Ghosh, S. K., Kalita, J. M., Ginjupalli, M. C. and K, K. R. 2020. Discovery of potential 1,3,5-Triazine compounds against strains of Plasmodium falciparum using supervised machine learning models. European Journal of Pharmaceutical Sciences, 144, 1-8.
Samarasinghe, T., Gunawardena, T., Mendis, P., Sofi, M. and Aye, L. 2019. Dependency Structure Matrix and Hierarchical Clustering based algorithm for optimum module identification in MEP systems. Automation in Construction, 104, 153-178.
Sheshukov, A. Y., Douglas-Mankin, K. R., Sinnathamby, S. and Daggupati, P. 2016. Pasture BMP effectiveness using an HRU-based subarea approach in SWAT. Journal of Environmental Management, 166, 276-284.
Song, L. R. and Zhang, J. Y. 2012. Hydrological Response to Climate Change in Beijiang River Basin Based on the SWAT Model. 2012 International Conference on Modern Hydraulic Engineering, 28, 241-245.
Tang, C., Bian, M., Liu, X., Li, M., Zhou, H., Wang, P. and Yin, H. 2019. Unsupervised feature selection via latent representation learning and manifold regularization. Neural Networks, 117, 163-178.
Usama, M., Qadir, J., Raza, A., Arif, H., Yau, K. L. A., Elkhatib, Y., Hussain, A. and Al-Fuqaha, A. 2019. Unsupervised Machine Learning for Networking: Techniques, Applications and Research Challenges. Ieee Access, 7, 65579-65615.
Wang, B., Ning, L. J. and Kong, Y. 2019. Integration of unsupervised and supervised machine learning algorithms for credit risk assessment. Expert Systems with Applications, 128, 301-315.
Wangpimool, W., Pongput, K., Sukvibool, C., Sombatpanit, S. and Gassman, P. W. 2013. The effect of reforestation on stream flow in Upper Nan river basin using Soil and Water Assessment Tool (SWAT) model. International Soil and Water Conservation Research, 1(2), 53-63.
Weyenberg, G. and Yoshida, R. (2015). Chapter 12 - Reconstructing the Phylogeny: Computational Methods. Algebraic and Discrete Mathematical Methods for Modern Biology. R. S. Robeva. Boston, Academic Press: 293-319.
Wu, J., Yen Primary, H., Arnold, J. G., Yang, Y.-C. E., Cai, X., White, M. J., Santhi, C., Miao, C. and Srinivasan, R. 2020. Development of Reservoir Operation Functions in SWAT+ for National Water Quantity and Quality Assessments. Journal of Hydrology, 583, 1-21.
Yang, J., Grunsky, E. and Cheng, Q. M. 2019. A novel hierarchical clustering analysis method based on Kullback-Leibler divergence and application on dalaimiao geochemical exploration data. Computers & Geosciences, 123, 10-19.
Zambelli, A. E. 2016. A data-driven approach to estimating the number of clusters in hierarchical clustering. F1000Res, 5, 1-6.
Zhou, F., Xu, Y. P., Chen, Y., Xu, C. Y., Gao, Y. Q. and Du, J. K. 2013. Hydrological response to urbanization at different spatio-temporal scales simulated by coupling of CLUE-S and the SWAT model in the Yangtze River Delta region. Journal of Hydrology, 485, 113-125.
Zou, Q., Cui, P., He, J., Lei, Y. and Li, S. S. 2019. Regional risk assessment of debris flows in China-An HRU-based approach. Geomorphology, 340, 84-102.
1 - https://orange3-imageanalytics.readthedocs.io/en/latest/widgets/imageembedding.html (02.01.2020).
2 - http://docs.biolab.si/3/visual-programming/widgets/unsupervised/distances.html (03.01.2020).
3 - https://orange.biolab.si (20.09.2020).