Türkiye’nin İllere Göre Konut Satış Miktarının Sinir Ağları Yöntemiyle İncelenmesi

Yıl 2025, Cilt: 15 Sayı: 2, 229 - 245, 27.10.2025

Şeyma Nur Ünal

Öz

Konut sektörü bir ülkenin sosyal ve ekonomik kalkınmasında çok önemli bir rol oynamaktadır. Sektör, istihdam, tasarruf, yatırım ve işgücü verimliliği gibi başlıca makroekonomik göstergeler üzerindeki etkisiyle ekonomik büyüme ve kalkınmaya katkıda bulunmaktadır. Bilişim teknolojisinin ilerlemesiyle birlikte, makine öğreniminin konut sektöründe, özellikle de konut satış analizi için uygulanması giderek daha önemli hale gelmiştir. Bu araştırma illere göre konut satış miktarını Yapay Sinir Ağları (YSA) yöntemiyle analiz etmeyi amaçlamaktadır. Kullanılan veriler, sanayi üretim endeksi, tüketici güven endeksi, inşaat güven endeksi, ekonomik güven endeksi, döviz kurudur ve illere göre konut satış miktarıdır. Algoritmaların analizi, her iki algoritmanın performans test sonuçlarının Ortalama Karesel Hata (OKH), Kök Ortalama Karesel Hata (KOKH) ve R-Kare (R2) gibi regresyon modelleri için performans metrikleri kullanılarak karşılaştırılmasıyla gerçekleştirilir. Ek olarak, bu araştırma eğitim, doğrulama ve test verileri arasında hangi veri oranının en iyi sonuçları verdiğini analiz eder. Araştırma bulguları, OKH 0.0011 ve KOKH 0.0331 değerleri performans ölçütü ileri beslemeli sinir ağının 10-2 ağa sahip modelin YSA içinde en iyi performansı ürettiğini göstermektedir.

Anahtar Kelimeler

Makro İktisat , Konut Satışı , Yapay Zekâ , Makine Öğrenimi , Yapay Sinir Ağları

Analysis of Housing Sales Quantity in Turkey by Provinces Using Neural Network Method

Öz

The housing sector plays a very important role in the social and economic development of a country.The sector contributes to economic growth and development through its impact on major macroeconomic indicators such as employment, savings, investment and labor productivity.With the advancement of information technology, the application of machine learning in the housing industry, especially for housing sales analysis, has become increasingly important.This research aims to analyze the housing sales amount by province using Artificial Neural Networks (ANN) method.The data used are industrial production index, consumer confidence index, construction confidence index, economic confidence index, exchange rate and housing sales amount by province.Analysis of the algorithms is performed by comparing the performance test results of both algorithms using performance metrics for regression models such as Mean Square Error (MSE), Root Mean Square Error (RMSE) and R-Square (R2).Additionally, this research analyzes which data ratio among training, validation, and test data yields the best results.Research findings show that the performance criterion of feed forward neural network with 10-2 network produces the best performance within the ANN with MSE 0.0011 and RMSE 0.0331 values.

Anahtar Kelimeler

Macroeconomics , Housing Sale , Machine Learning , Artificial Neural Network , Artificial Intelligence

Kaynakça

• Abidoye, R. B., ve Chan, A. P. (2017). Artificial neural network in property valuation: Application Framework and Research. Property Management, 5(35), 554-571.
• Ahmad, M. W., Mourshed, M. ve Rezgui, Y. (2017). Trees vs neurons: comparison between random forest and ann for high-resolution prediction of building energy consumption. Energy and Buildings (147), 77-69.
• Ahmed, S. F., Alam, M. S., Hassan, M., Rozbu, R. M., Ishtiak, T., Rafa, N., vd. (2023). Deep Learning Modelling Techniques: Current Progress, Applications, Advantages, and Challenges. Artifcial Intelligence Review (56), 13521–13617.
• Bai, H., ve Chen, X. (2023). House price forecasting in ames based on bayesian regularized BP neural network. Automation and Machine Learning, 4(1), 17-23.
• Capital, F. (2024, 06 19). https://fastercapital.com/topics/understanding-the-housing-industry-and-its-impact-on-the-economy.html
• Chan, Z. S., Ngan, H. W., ve Rad, A. B. (2003). Improving bayesian regularization of ANN via pre-training with early-stopping. Neural Processing Letters, 1(18), 29-34.
• Çavuşlu, M. A., Becerikli, Y., ve Karakuzu, C. (2012). Levenberg – Marquardt algoritması ile YSA eğitiminin donanımsal gerçeklenmesi. TBV Bilgisayar Bilimleri ve Mühendisliği Dergisi, 5, 31-38.
• Elshamy, M. M., Tiraturyan, A. N., Uglova, E. V., ve Elgendy, M. Z. (2021). Comparison of Feed-Forward, Cascade-Forward, and Elman algorithms models for determination of the elastic modulus of pavement layers. (s. 46-53). ICGDA.
• Erdinç, M. H. (1990). Türkiye'de konut sektörünün ekonomik analizi (1979-1988) (Yüksek lisans tezi). Anadolu Üniversitesi Sosyal Bilimler Enstitüsü.
• Ersöz, S., Yaman, N., ve Birgören, B. (2008). Müşteri ilişkileri yönteminde verilerin yapay sinir ağları ile modellenmesi ve analizi. Gazi Üniv. Müh. Mim. Fak. Dergisi, 23(4), 759-767.
• Ghorbani, S. ve Afgheh, S. M. (2017). Forecasting the house price for ahvaz city: the comparison of. Urban Economics and Management, 3(5), 29-45. Gogas, P. ve Papadimitriou, T. (2021). Machine learning in economics and finance. Computational Economics (57), 1-4. Gurney, K. (1997). An Introduction To Neural Networks. UCL Press.
• Hodson, T. O. (2022). Root mean squre error (RMSE) or Mean absolute error (MAE): When to use them or not. Geoscientific Model Development, 14(15), 5481-5487. https://doi.org/10.5194/gmd-15-5481-2022
• Ibtissem, C. ve Nouredine, L. (2013). A Hybrid method based on conjugate gradient trained neural network and differential evolution for non linear systems ıdentification. ICEESA. Hammamet, Tunisia.
• Lee, H., Han, H., Pettit, C., Gao, Q., ve Shi, V. (2024). Machine learning approach to residential valuation: a convolutional neural network model for geographic variation. The Annals of Regional Science (72), 579-599.
• Nazemi, B. ve Rafiean, M. (2022). Modelling the affecting factors of housing price using gmdh-type artificial neural networks in Isfahan city of Iran. International Journal of Housing Markets and Analysis, 15(1), 4-18.
• Nazemi, B. ve Rafiean, M. (2022). Modelling the affecting factors of housing price using GMDH-type artificial neural networks in Isfahan city of Iran. International Journal of Housing Markets and Analysis, 15(1), 4-18.
• Obaido, G., Mienye, I. D., Oluwaseun, E. F., Emmanuel, D. I., Ogunleye, A., Ogbuokiri, B. vd. (2024). Supervised machine learning in drug discovery and development: algorithms, applications, challenges, and prospects. Machine Learning with Applications (17).
• Oshodi, O. S., Thwala, W. D., Odubiyi, T. B., Abidoye, R. B.,ve Aigbavboa, C. O. (2019). Using neural network model to estimate the rental price of residential properties. Journal of Financial Management of Property and Construction, 2(12).
• Park, B., ve Bae, J. K. (2015). Using machine learning algorithms for housing price prediction: the case of fairfax county, virginia housing data. Expert Systems with Applications, 6(42), 2928-2934.
• Rampini, L. ve Cecconi, F. (2021). Artificial intelligence algorithms to predict Italian real estate market prices. Journal of Property Investment & Finance, 6(40), 588-611.
• Sa’at, N. F., Maimun, N. H. ve Idris, N. H. (2021). Enhancing the accuracy of Malaysian house price forecastıng: a comparative analysis on the forecasting performance between the hedonıc prıce model and artificial neural network model. Planning Malaysia Journal, 3(19), 249-259.
• Sathiavelu, S. K. ve Mouleswaran, S. (2024). Investigation of machine faults using elman neural network and decision tree. ICAMDMS. EAI.
• Soltani, A. ve Lee, L. C. (2024). The non-linear dynamics of south australian regional housing markets: A Machine Learning Approach. Applied Geography (166).
• Song, Z., Sun, F., Zhang, R., Du, Y., ve Li, C. (2021). Prediction of road network traffic state using the NARX neural network. Advanced Transportation, 774-786.
• Tekouabou, S. C., Gherghina, Ş. C., Kameni, E. D., Filali, Y. ve Gartoumi, K. I. (2023). AI-based on machine learning methods for urban real estate prediction: A systematic survey. Archives of Computational Methods in Engineering (31), 1079-1095.
• Winky, K. O., Tang, B.-S. ve Wong, S. W. (2020). Predicting property prices with machine learning algorithms. Journal of Property Research, 1, 48-70.
• Xu, X., ve Zhang, Y. (2024). Office property price ındex forecasting using neural networks. Journal of Financial Management of Property and Construction, 29(1), 52-82.
• Yu, H., ve Wilamowski, B. M. (2011). Levenberg–Marquardt training. Auburn University: https://www.eng.auburn.edu/~wilambm/pap/2011/K10149_C012.pdf