Estimation of Air Temprerature Using Geographically and Altitudinal Weighted Regression

Year 2020, Ejosat Special Issue 2020 (HORA), 81 - 86, 15.08.2020

Murat Taşyürek , Mete Çelik

https://doi.org/10.31590/ejosat.779074

Cited By: 2

Abstract

The interest in location-based analysis has increased with the collection of data in different application domains using the internet of things technologies. Geographically Weighted Regression (GWR), one of the commonly used location-based analysis methods, is a local spatial regression technique that models changing relationships on geography. Geographically and Altitudinal Weighted Regression (GAWR) is an approach developed by adding altitude relationships to the GWR approach. The GAWR model does modeling by considering the spatial (horizontal) and altitude (vertical) information in the data, and therefore it can produce successful results in datasets (such as meteorological datasets) that have spatial and altitude relationships. In the literature, temperature estimation studies have been done by using GWR. This study proposes to use the GAWR algorithm to estimate the air temperature. In the study, the meterlogical dataset, taken from the General Directorate of Meteorology (MGM), was used and altitude, pressure, and humidity information was used to estimate temperature value of a location. Results show that GAWR outperforms GWR algorithm.

Keywords

Geographically and altidudinal weighted regression, Altitudinal weight, Estimation of Air Temprature, Data mining

Hava Sıcaklık Değerlerinin Coğrafi ve Rakım Ağırlıklı Regresyon Yöntemi ile Tahmin Edilmesi

Abstract

Nesnelerin interneti sayesinde farklı alanlarda çok fazla sayıda ve çeşitte mekânsal verinin toplanması konum temelli analizlere olan ilgiyi arttırmıştır. Yaygın olarak kullanılan konum temelli analiz yöntemlerininden birisi olan coğrafi ağırlıklı regresyon (Geographically Weighted Regression - GWR), coğrafya üzerindeki değişen ilişkileri modelleyen bir yerel mekânsal regresyon tekniğidir. Coğrafi ve rakım ağırlıklı regresyon (Geographically and Altitudinal Weighted Regression - GAWR) ise GWR yaklaşımına rakım (yükseklik) ilişkilerin eklenmesiyle geliştirilen bir yaklaşımdır. GAWR, GWR modelinden farklı olarak, verideki mekân (yatay) ve rakım (dikey) bilgilerini gözönüne alarak modelleme yapar ve bu nedenle, hem mekân ve hem de rakım ilişkilerinin olduğu veri kümelerinde (meteorolojik veri kümeleri gibi) başarılı sonuçlar verebilmektedir. Literatürde GWR ile sıcaklık tahmin çalışmaları yapılmıştır. Bu çalışmada, sıcaklık tahmini için GAWR algoritmasının kullanılması önerilmiştir. Çalışmada Meteoroloji Genel Müdürlüğü’nden (MGM) alınan veriler kullanılmıştır ve bir noktanın sıcaklıkğının tahmini için o noktanın rakım, basınç ve nem bilgisi kullanılmıştır. Sonuçlar GAWR algoritmasının GWR algoritmasına göre daha doğru sonuçlar ürettiğini göstermiştir.

Keywords

Coğrafi ve rakım ağırlıklı regresyon, Dikey ağırlık, Hava sıcaklık tahmini, Veri madenciliği

References

• Brook, R. J., & Arnold, G. C. (2018). Applied regression analysis and experimental design. Routledge.
• Celik, M., Kazar, B. M., Shekhar, S., & Boley, D. (2006). Parameter Estimation for the Spatial Autoregression Model: A Rigorous Approach£. In Proceedings of the 2nd NASA Data Mining Workshop: Issues and Applications in Earth Science with the 38th Symposium on the Interface of Computing Science, Statistics and Applications.
• Cho, S. H., Lambert, D. M., & Chen, Z. (2010). Geographically weighted regression bandwidth selection and spatial autocorrelation: an empirical example using Chinese agriculture data. Applied Economics Letters, 17(8), 767-772.
• Dadaser-Celik F., Celik M., Dokuz A. (2012). Associations between stream flow and climatic parameters at Kızılırmak River Basin in Turkey. Global Nest Journal, 14, 354-361.
• Fotheringham, A. S., Brunsdon, C., & Charlton, M. (2003). Geographically weighted regression: the analysis of spatially varying relationships, John Wiley & Sons.
• Fotheringham, A. S., Yang, W., & Kang, W. (2017). Multiscale geographically weighted regression (MGWR). Annals of the American Association of Geographers, 107(6), 1247-1265.
• Gollini, I., Lu, B., Charlton, M., Brunsdon, C., & Harris, P. (2013). GWmodel: an R package for exploring spatial heterogeneity using geographically weighted models. arXiv preprint arXiv:1306.0413.
• Guo, L., Ma, Z., & Zhang, L. (2008). Comparison of bandwidth selection in application of geographically weighted regression: a case study. Canadian Journal of Forest Research, 38(9), 2526-2534.
• Kauermann, G., & Opsomer, J. D. (2004). Generalized cross-validation for bandwidth selection of backfitting estimates in generalized additive models. Journal of Computational and Graphical Statistics, 13(1), 66-89.
• Kayseri Büyükşehir Belediyesi. (2020, March 10). Retrieved from https://www.kayseri.bel.tr.
• Kazar, B. M., & Celik, M. (2012). Spatial Autoregression (SAR) Model: Parameter Estimation Techniques, Springer Briefs in computer Science, ISBN:978-1461418412, Springer.
• Li, Z., Fotheringham, A. S., Li, W., & Oshan, T. (2019). Fast Geographically Weighted Regression (FastGWR): a scalable algorithm to investigate spatial process heterogeneity in millions of observations. International Journal of Geographical Information Science, 33(1), 155-175.
• Lu, B., Brunsdon, C., Charlton, M., & Harris, P. (2017). Geographically weighted regression with parameter-specific distance metrics. International Journal of Geographical Information Science, 31(5), 982-998.
• Lu, B., Yang, W., Ge, Y., & Harris, P. (2018). Improvements to the calibration of a geographically weighted regression with parameter-specific distance metrics and bandwidths. Computers, Environment and Urban Systems, 71, 41-57.
• Novo, O. (2018). Blockchain meets IoT: An architecture for scalable access management in IoT. IEEE Internet of Things Journal, 5(2), 1184-1195.
• Pabuçcu, A. (2016). Fen bilgisi öğretmen adaylarının gaz basıncıyla ilgili bilgilerini günlük hayatla ilişkilendirebilme seviyeleri. Turkiye Kimya Dernegi Dergisi Kisim C: Kimya Egitimi, 1(2), 1-24.
• Perera, C., Zaslavsky, A., Christen, P., & Georgakopoulos, D. (2014). Sensing as a service model for smart cities supported by internet of things. Transactions on emerging telecommunications technologies, 25(1), 81-93.
• Prasad, A. V. (2017). Exploring the Convergence of Big Data and the Internet of Things, IGI Global.
• Propastin, P. (2012). Modifying geographically weighted regression for estimating aboveground biomass in tropical rainforests by multispectral remote sensing data. International Journal of Applied Earth Observation and Geoinformation, 18, 82-90.
• Shekhar, S., Vatsavai, R.R., & Celik, M. (2009). Spatial and Spatiotemporal Data Mining: Recent Advances, as a chapter of Next Generation of Data Mining, H. Kargupta, J. Han, P.S. Yu, R. Motwani, and Vipin Kumar (Eds.), ISBN:978-1-4200-8586-0, CRC Press
• Stergiou, C., Psannis, K. E., Kim, B. G., & Gupta, B. (2018). Secure integration of IoT and cloud computing. Future Generation Computer Systems, 78, 964-975.
• Tasyurek, M., & Celik, M. (2020). RNN-GWR: A geographically weighted regression approach for frequently updated data, Neurocomputing, 399, 258-270.
• Taşyürek, M., & Çelik, M. (2020). Akıllı Durak Sistemindeki Araç Seyahat Sürelerinin Birleşik Yapay Sinir Ağları Kullanarak Tahmini. Avrupa Bilim ve Teknoloji Dergisi, 72-79. DOI: 10.31590/ejosat.araconf10.
• Tobler, W. R. (1970). A computer movie simulating urban growth in the Detroit region. Economic geography, 46(sup1), 234-240.
• Yıldırım, G., & Tatar, Y. (2019). Uzak kullanıcı destekli bir IoT-WSN sanal laboratuvarı ve test platformu: FıratWSN. Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi, 34(4), 1831-1846.
• Zougab, N., Adjabi, S., & Kokonendji, C. C. (2014). Bayesian estimation of adaptive bandwidth matrices in multivariate kernel density estimation. Computational Statistics & Data Analysis, 75, 28-38.