Mesken nitelikli gayrimenkul fiyat tahmininde ağaç tabanlı makine öğrenmesi algoritmalarının karşılaştırılması

Year 2024, , 116 - 130, 15.03.2024

Ayşe Yavuz Özalp , Halil Akıncı

https://doi.org/10.17714/gumusfenbil.1363531

Abstract

Mesken nitelikli gayrimenkuller, güvenli ve karlı bir yatırım aracı olarak kabul edilirken aynı zamanda temel insan hakkı olan barınma ihtiyacını da karşılamaktadır. Konutun hem değerini etkileyen hem de yere, kişiye ve zamana göre değişen çok sayıda parametrenin varlığı değerleme sürecini zorlaştırmaktadır. Bu bakımdan, doğru ve gerçekçi fiyat tahmini başta alıcılar olmak üzere sektörün tüm paydaşları için büyük önem taşımaktadır. Klasik matematiksel modelleme yöntemlerine alternatif olan makine öğrenme algoritmaları, fiyat tahmin modellerinin etkenliğini ve başarısını artırma bağlamında önemli olanaklar sunmaktadır. Dolayısıyla bu çalışmanın amacı, Artvin Kent Merkezinde ağaç temelli makine öğrenme algoritmalarından Rastgele Orman (RF), Gradyan Artırma Makinaları (GBM), AdaBoost ve Aşırı Gradyan Artırma (XGBoost) yöntemlerinin konut değerlemede uygulanabilirlikleri ve tahmin performanslarının araştırılmasıdır. Çalışmanın sonucunda, XGBoost ve RF algoritmaları, konut değerini tahmin etmede Korelasyon Katsayısı (Correlation Coefficients- R2), Ortalama Mutlak Hata (Mean Absolute Error - MAE) ve Karesel Ortalama hata (Root Mean Squared Error- RMSE) ölçütlerinin tümünde (sırasıyla 0.705 ve 0.701) en iyi performansı göstermiştir. Böylece, başta XGBoost ve RF olmak üzere makine öğrenme algoritmalarının konut değerlemede az veri durumunda bile yeterli performans gösterdiği, veri setinin genişletilmesiyle birlikte başarı performansının daha da artacağı söylenebilir.

Keywords

AdaBoost, GBM, RF, Mesken nitelikli gayrimenkul, Değerleme, XGBoost

Comparison of tree-based machine learning algorithms in price prediction of residential real estate

Abstract

Residential real estate is regarded as a safe and profitable investment tool while also meeting the basic human right to housing. The fact that there exists a large number of parameters both affecting the value of a house and varying based on place, person, and time makes the valuation process difficult. In this regard, accurate and realistic price prediction is critical for all stakeholders, particularly purchasers. Machine learning algorithms as an alternative to classical mathematical modeling methods offer great prospects for boosting the efficacy and success rate of price estimating models. Therefore, the purpose of this study is to investigate the applicability and prediction performance of the tree-based ML algorithms -Random Forest (RF), Gradient Boosting Machine (GBM), AdaBoost, and Extreme Gradient Boosting (XGBoost)- in house valuation for Artvin City Center. As a result of the study, the XGBoost and RF algorithms performed the best in estimating house value (0.705 and 0.701, respectively) as determined by the Correlation Coefficients (R2), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE) metrics. Thus, it can be said that ML algorithms, particularly XGBoost and RF, perform satisfactorily in residential real estate appraisal even with modest amounts of data and that the success rate grows as the amount of data increases.

Keywords

AdaBoost, GBM, RF, Residential real estate, Valuation, XGBoost

References

• Adetunji, A.B., Akande, N., Ajala, F.A., Oyewo, O., Akande, Y.F., & Oluwadara, G. (2022). House price prediction using random forest machine learning technique. Procedia Computer Science, 199, 806–813. https://doi.org/10.1016/j.procs.2022.01.100
• Afonso, B.K.A., Melo, L.C., Oliveira, W.D.G., Sousa, S.B.S., & Berton, L. (2019). Housing prices prediction with a deep learning and random forest ensemble. Anais do Encontro Nacional de Inteligência Artificial e Computacional (ENIAC 2019) (pp. 389-400), Salvador.
• Afşar, M., & Yüksel, Ö.G. (2022). The effectiveness of the housing channel in monetary policy. ESOGU Journal of Economics and Administrative Sciences, 17(2), 345 – 367. https://doi.org/10.17153/oguiibf.1064471
• Akay, E.C., Topal, K.H., Kizilarslan, S., & Bulbul, H. (2019). Forecasting of Turkish housing price index: ARIMA, random forest, ARIMA-random forest. PressAcademia Procedia, 10, 7-11. https://doi.org/10.17261/Pressacademia.2019.1134
• Akinci, H. (2022). Assessment of rainfall-induced landslide susceptibility in Artvin, Turkey using machine learning techniques. Journal of African Earth Sciences, 191, 104535. https://doi.org/10.1016/j.jafrearsci.2022.104535
• Alkan, T., Dokuz, Y., Ecemiş, A., Bozdağ, A., & Durduran, S. (2022). Using machine learning algorithms for predicting real estate values in tourism centers. Data Analytics and Machine Learning, 27, 2601–2613. https://doi.org/10.1007/s00500-022-07579-7
• Antipov, E.A., & Pokryshevskaya, E.B. (2012). Mass appraisal of residential apartments: An application of random forest for valuation and a CART-based approach for model diagnostics. Expert Systems with Applications, 39, 1772-1778. https://doi.org/10.1016/j.eswa.2011.08.077
• Arslan, Y., Ceritoğlu, E., & Kanık, B. (2022, October 14). The effects of demographic changes on the long-term housing demand in Turkey. Munich Personal Repec Archive. https://mpra.ub.uni-muenchen.de/52013/
• Avcı, C., Budak, M., Yagmur, N., & Balcık, F. B. (2023). Comparison between random forest and support vector machine algorithms for LULC classification. International Journal of Engineering and Geosciences, 8(1), 01-10. https://doi.org/10.26833/ijeg.987605
• Aydemir, E., Aktürk, C., & Yalçınkaya, M.A. (2020). Estimation of housing prices with artificial intelligence. Turkish Studies, 15(2), 183-194. http://dx.doi.org/10.29228/TurkishStudies.43161
• Aydinoglu, A.C., Bovkir, R., & Colkesen, I. (2021). Implementing a mass valuation application on interoperable land valuation data model designed as an extension of the national GDI. Survey Review, 53, 349-365. https://doi.org/10.1080/00396265.2020.1771967
• Baldominos, A., Blanco, I., Moreno, A., Iturrarte, R., Bernardez, O., & Afonso, C. (2018). Identifying real estate opportunities using machine learning. Applied Sciences, 8(11), 2321. https://doi.org/10.48550/arXiv.1809.04933
• Başer, U., & Bozoğlu, M. (2019). Determination of the factors affecting housing rent using hedonic price model: the case of Ilkadım and Atakum districts of Samsun province. Eurasian Journal of Researches in Social and Economics, 6(4), 308-316.
• Bilgilioğlu, S.S., & Yılmaz, H.M. (2021). Comparison of different machine learning models for mass appraisal of real estate. Survey Review, 55, 32-43. https://doi.org/10.1080/00396265.2021.1996799
• Borst, R.A. (1991). Artificial neural networks: the next modelling/calibration technology for the assessment community. Property Tax Journal, 10(1), 69–94.
• Breiman, L. (2001). Random Forests. Machine Learning, 45(l), 5–32.
• Büyük, G., & Ünel, F. B. (2021). Comparison of modern methods using the python programming language in mass housing valuation. Advanced Land Management, 1(1), 23-31.
• Can, R., Kocaman, S., & Gokceoglu, C. (2021). A comprehensive assessment of XGBoost algorithm for landslide susceptibility mapping in the upper basin of Ataturk Dam, Turkey. Applied Science, 11, 4993. https://doi.org/10.3390/app11114993
• Ceh, M., Kilibarda, M., Lisec, A., & Bajat, B. (2018). Estimating the performance of random forest versus multiple regression for predicting prices of the apartments. ISPRS International Journal of Geo-Information, 7(5), 168. https://doi.org/10.3390/ijgi7050168
• Chen, T., & Guestrin, C. (2016). XG Boost: A scalable tree boosting system. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining-KDD (pp. 785–794), New York.
• Chen, W., Xie, X., Wang, J., Pradhan, B., Hong, H., Tien, B., Duan, Z., & Ma, J. (2017). A comparative study of logistic model tree, random forest, and classification and regression tree models for spatial prediction of landslide susceptibility. Catena, 151, 147-160. https://doi.org/10.1016/j.catena.2016.11.032
• Corine- Coordination of Information on the Environment. (2022, November 29). https://corinecbs.tarimorman.gov.tr/ corine
• Doğan, O., Bande, N., Genç Y., & Akyön, F.Ç. (2022). Estimation of housing fair values using artificial neural networks method in Kecioren/Ankara. International Journal of Economic and Administrative Studies, 35, 113-128. https://doi.org/10.18092/ulikidince.941952
• Embaye, W.T., Zereyesus, Y.A., & Chen, B. (2021). Predicting the rental value of houses in household surveys in Tanzania, Uganda and Malawi: Evaluations of hedonic pricing and machine learning approaches. Plos One. 16, 1-20. https://doi.org/10.1371/journal.pone.0244953
• Esen, Y., & Tokgöz, H. (2021). A different perspective to real estate valuation with fuzzy logic modeling. Journal of Engineering Sciences and Design, 9(4), 1155-1165. https://doi.org/10.21923/jesd.876523
• Fei, Y. (2020). California rental price prediction using machine-learning algorithms [Master’s Thesis, University of California Center for Social Statistics].
• Freund, Y., & Schapire, R.E. (1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 55(1), 119–139.
• Friedman, J.H. (2001). Greedy function approximation: a gradient boosting machine. The Annals of Statistics, 29(5), 1189–1232. https://doi.org/10.1214/aos/1013203451
• GDLRC- General Directorate of Land Registry and Cadastre. (2022, September 10). https://parselsorgu.tkgm.gov.tr/
• Gustafsson, A., & Wogenius, S. (2014). Modelling apartment prices with the multiple linear regression model. Royal Institute of Technology. https://www.diva-portal.org/smash/get/diva2:725045/FULLTEXT01.pdfMultiple
• Hayrullahoğlu, G., Aliefendioğlu, Y., Tanrıvermiş, H., & Hayrullahoğlu, A.C. (2018). Estimation of the hedonic valuation model in housing markets: the case of Cukurambar region in Çankaya district of Ankara province. Ecoforum, 1, 1-9.
• He, Q., Jiang, Z., Wang, M., & Liu, K. (2021). Landslide and wildfire susceptibility assessment in Southeast Asia using ensemble machine learning methods. Remote Sensing, 13(8), 1572. https://doi.org/10.3390/rs13081572
• Hjort, A., Pensar, J., Scheel, I., & Sommervoll, D.E. (2022). House price prediction with gradient boosted trees under different loss functions. Journal of Property Research, 39(4), 338-364. https://doi.org/10.1080/09599916.2022.2070525
• Hong, J., Choi, H., & Kim, W.S. (2020). A house price valuation based on the random forest approach: the mass appraisal of residential property in South Korea. International Journal of Strategic Property Management, 24(3), 140–152. https://doi.org/10.3846/ijspm.2020.11544
• IAAO- International Association of Assessing Officers. (2022, April 14). https://www.iaao.org/media/standards/ StandardOnMassAppraisal.pdf
• Iban, M.C. (2021). Accuracy analysis of community algorithms in estimating the value of real estate. 1st International Artificial Intelligence and Data Science Congress (pp. 226-232), İzmir.
• İlhan, A.T., & Öz, N.S. (2020). Applicability of artificial neural networks on mass valuation of real estates: the case of Gölbaşı District. Hacettepe University Journal of Social Sciences, 2(2), 160-188.
• Louati, A., Lahyani, R., Aldaej, A., Aldumaykhi, A., & Otai, S. (2021). Price forecasting for real estate using machine learning: A case study on Riyadh city. Concurrency and Computation: Practice and Experience, 1-16. https://doi.org/10.1002/cpe.6748
• Mete, M.O., & Yomralıoğlu, T. (2022). GIS and machine learning based mass valuation of residential properties. 11th Turkish National Photogrammetry and Remote Sensing Union (TUFUAB) Technical Symposium (pp. 1-5), Mersin.
• Oral, M., Okatan, E., & Kırbaş, İ. (2021). A study on house price prediction using machine learning methods: the case of Madrid. 3rd International young researchers student congress (pp. 263-272), Burdur.
• Özalp, A.Y., Akıncı, H., & Temuçin Kılıçer, S. (2020). Analysis of parameters affecting value of real estates with land property in Artvin Case. Geomatik, 5(2), 100-111. https://doi.org/10.29128/geomatik.579401
• Ozdemir, M., Yıldız, K., & Büyüktanır, B. (2022). Housing price estimation with deep learning: a case study of Sakarya Turkey. BSEU Journal of Science, 9(1), 138-151. https://doi.org/10.35193/bseufbd.998331
• Ravikumar, A.S. (2016). Real estate price prediction using machine learning [Master’s Thesis, National College of Ireland School of Computing].
• Saraç, E. (2012). Real estate appraisal with artificial neural networks method [Master’s Thesis, İstanbul Kültür University Institute of Science].
• Sahin, E.K. (2020). Assessing the predictive capability of ensemble tree methods for landslide susceptibility mapping using XGBoost, gradient boosting machine, and random forest. SN Applied Science, 2,1308. https://doi.org/10.1007/s42452-020-3060-1
• Schapire, R.E. (2013). Explaining AdaBoost. In Empirical Inference (Bernhard Schölkopf, Zhiyuan Luo, Vladimir Vovk Eds.). Springer Berlin.
• Sevgen, S.C., & Tanrivermis, Y.A. (2020). Mass apprasial with a machine learning algorithm: random forest regression. Journal of Information Technologies, 13(3), 301-311. https://doi.org/10.17671/gazibtd.555784
• Tabar, M.E., Başaran, A.C., & Şişman, Y. (2021). Housing valuation study in Tokat province with multiple regression and artificial neural networks. Turkish Journal of Land Management, 3(1), 01-07. https://doi.org/10.51765/tayod.832227
• Tabanoğlu, M. (2019). Estimating the market value of residential buildings with artificial neural networks method: Düzce sample [Master’s Thesis, Düzce University Graduate School of Natural Sciences].
• TDUB-Türkiye Değerleme Uzmanları Birliği. (2022, April 15). https://tdub.org.tr/uploads/ documents/1667807030_d7767c42b3a070c20179
• TSI- The Turkish Statistical Institute. (2022, November 10). https://data.tuik.gov.tr/Kategori/GetKategori?p=Insaat-ve-Konut-116.
• Tuna, M.F., Türk, T., & Kitapçı, O. (2015). House prices with the help of linear regression and GIS estimating: the example of Ankara. TMMOB HKMO 15. Türkiye Harita Bilimsel ve Teknik Kurultayı (pp. 1-5), Ankara.
• Ulvi, C., & Özkan, G. (2019). Usability of artificial intelligence techniques at real estate valuation and comparison of the methods. Journal of Geomatics, 4(2), 134-140.
• Wang, C., & Wu, H. (2018). A new machine learning approach to house price estimation. New Trends in Mathematical Sciences, 6(4),165-171. https://doi.org/10.20852/ntmsci.2018.327
• Wang, Z., Liu, Q., & Liu, Y. (2020). Mapping landslide susceptibility using machine learning algorithms and GIS: A case study in Shexian County, Anhui Province, China. Symmetry, 12, 1954. https://doi.org/10.3390/sym12121954
• Wilkowski, W., & Budzynski, T. (2006). Application of artificial neural networks for real estate valuation. Shaping the Change XXIII FIG Congress (pp. 1-12), Munich.
• Wu, Y., Ke, Y., Chen, Z., Liang, S., Zhao, H., & Hong, H. (2020). Application of alternating decision tree with AdaBoost and bagging ensembles for landslide susceptibility mapping. Catena, 187, 104396. https://doi.org/10.1016/j.catena.2019.104396
• XGBoost- XGBoost Python Package. (2022, December 13). https://xgboost. readthedocs.io/en/stable/python/
• Yavuz Ozalp, A., & Akinci, H. (2017). The use of hedonic pricing method to determine the parameters affecting residential real estate prices. Arabian Journal of Geoscience, 10, 535. https://doi.org/10.1007/s12517-017-3331-3
• Yavuz Özalp, A., & Akıncı, H. (2018). Using hedonic pricing model to analyze parameters affecting residential real estate value in Artvin City Center. FIG Congress 2018 (pp. 1-15), İstanbul.
• Yazdani, M. (2021). Machine learning, deep learning, and hedonic methods for real estate price prediction [Master’s Thesis, Colorado University Department of Economics].
• Yıldırımer, S., Özalp, M., & Erdoğan Yüksel, E. (2015). Determining loss and degradation of lands as a result of large dam projects and associated road constructions within the Coruh River Watershed. ACU Journal of Forestry Faculty, 6(1), 1-17. https://doi.org/10.17474/acuofd.00766
• Yılmazel, Ö., Afşar, E., & Yılmazel, S. (2018). Using artificial neural network method to predict housing prices. International Journal of Economic and Administrative Studies, 20, 285-300. https://doi.org/10.18092/ulikidince.341584
• Yılmazer, S., & Kocaman, S. (2020). A mass appraisal assessment study using machine learning based on multiple regression and random forest. Land Use Policy, 99, 104889. https://doi.org/10.1016/j.landusepol.2020.104889
• Yoshida, T., & Seya, H. (2021). Spatial prediction of apartment rent using regression-based and machine learning-based approaches with a large dataset. The Journal of Real Estate Finance and Economics, 1-39. https://doi.org/10.1007/s11146-022-09929-6
• Yu, D., Wei, Y.D., & Wu, C. (2007). Modeling spatial dimensions of housing prices in Milwaukee, WI. Environment and Planning B: Urban Analytics and City Science, 34(6), 1085–1102. https://doi.org/10.1068/b32119
• Zaki, J., Nayyar, A., Dalal, S., & Ali, Z.H. (2022). House price prediction using hedonic pricing model and machine learning techniques. Concurrency and Computation: Practice and Experience, 34, 1-15. https://doi.org/10.1002/cpe.7342