EFFECT OF OUTLIER DETECTION TECHNIQUES IN MACHINE LEARNING ASSISTED MASS REAL ESTATE VALUATION

Abstract

In mass real estate valuation, detecting outliers is crucial for improving model performance. This study examined how different outlier detection techniques impact model performance when using machine learning algorithms. A dataset with 78 variables and 37,269 market samples, enriched with geographic dataset, was created for Tuzla and Pendik districts in Istanbul. Outliers were detected using Boxplot, MAD, and Z-Score techniques. The Boxplot detected the most outliers (3,810), while Z-Score (|Z| > 3) detected the fewest (943). After removing outliers, the dataset was split into training (70%) and test (30%) sets. Then, Random Forest (RF) models were trained, and performance was evaluated using MAE, MAPE, RMSE, and R² metrics. Methods have emerged from different perspectives. Z-Score (|Z| > 3) produced the best results in terms of absolute error and overall performance, achieving the lowest MAE (0.046849) and RMSE (0.065869). The Boxplot method showed consistent performance, with notably low MAPE (12.933) and strong explanatory power (R² = 0.698), making it a good general-purpose technique. MAD provides the highest R² (0.700), making it the most effective in explaining variance. These findings highlight that determining appropriate outlier detection methods can lead to more accurate and robust valuation models by optimizing datasets.

Keywords

• GIS
• Machine Learning
• Mass Real Estate Valuation
• Outlier Detection
• Random Forest.

Supporting Institution

This study was supported by The Scientific and Technological Research Council of Turkey (TÜBİTAK) with project number 122R021.

Project Number

122R021

Ethical Statement

The authors declare that this study was conducted in accordance with all relevant ethical standards.

Thanks

The authors would also like to acknowledge Endeksa for their valuable support in providing data for this study.

References

1. S. Enemark, “Building Land Information Policies,” in Proc. UN, FIG, PC IDEA Inter-regional Special Forum on The Building of Land Information Policies in the Americas, Aguascalientes, Mexico, Oct. 26–27, 2004.
2. M. C. Ban, “An explainable model for the mass appraisal of residences: The application of tree-based Machine Learning algorithms and interpretation of value determinants,” Habitat International, vol. 128, p. 102660, 2022.
3. S. Dogan, L. Genc, S. C. Yucebas, and S. Yalpir, “Analyzing the impact of the 2023 general elections on land prices using machine learning: a case study in Canakkale, Turkey,” Konya Journal of Engineering Sciences, vol. 13, no. 1, pp. 147–164, 2025.
4. S. C. Yucebas, S. Yalpir, L. Genc, and M. Dogan, “Price prediction and determination of the affecting variables of the real estate by using X-means clustering and CART decision trees,” Journal of Universal Computer Science, vol. 30, no. 4, p. 531, 2024.
5. N. Genc, H. E. Colak, and F. Ozbilgin, “Spatial performance approach to machine learning algorithms: A GIS-based comparison analysis for real estate valuation,” Transactions in GIS, vol. 29, no. 1, p. e13303, 2025.
6. S. Şişman, N. Tellioğlu, and A. Ç. Aydınoğlu, “Konumsal-zamansal taşınmaz değer değişim dinamiklerinin CBS ve istatistiksel analiz teknikleri ile incelenmesi,” Jeodezi ve Jeoinformasyon Dergisi, vol. 12, no. 1, pp. 58–76, 2025.
7. A. C. Aydinoglu and S. Sisman, “Comparing modelling performance and evaluating differences of feature importance on defined geographical appraisal zones for mass real estate appraisal,” Spatial Economic Analysis, vol. 19, no. 2, pp. 225–249, 2024.
8. F. B. Ünel and Ş. Yalpır, “Türkiye’de gayrimenkulların değerini etkileyen kriterlere yaklaşım,” Geomatik, vol. 4, no. 2, pp. 112–133, 2019.

9. M. Yılmaz, “Gayrimenkul değerleme yöntemleri ve bir uygulama,” M.Sc. thesis, Marmara Üniversitesi, İstanbul, Türkiye, 2019.
10. E. Deveci and İ. Yılmaz, “Coğrafi bilgi sistemleri yardımıyla taşınmaz mal değerlemesi: Afyonkarahisar il merkezi örneği,” Harita Teknolojileri Elektronik Dergisi, vol. 1, no. 1, pp. 33–47, 2009.
11. S. Yener, “Gayrimenkul değerlemesinde değer haritalarının oluşturulması ve haritalar yardımıyla gayrimenkul değerinin tespit edilmesi”, Master’s thesis, Sosyal Bilimler Enstitüsü, 2017.
12. B. Iglewicz and D. C. Hoaglin, How to Detect and Handle Outliers. Milwaukee, WI: Quality Press, 1993. V. Barnett and T. Lewis, Outliers in Statistical Data, 3rd ed. New York, NY: Wiley, 1994
13. K. Singh and S. Upadhyaya, “Outlier detection: Applications and techniques,” International Journal of Computer Science Issues, vol. 9, no. 1, p. 307, 2012.
14. S. Seo, “A review and comparison of methods for detecting outliers in univariate data sets,” M.S. thesis, University of Pittsburgh, Pittsburgh, PA, 2006.
15. C. H. Sim, F. F. Gan, and T. C. Chang, “Outlier labeling with boxplot procedures,” Journal of the American Statistical Association, vol. 100, no. 470, pp. 642–652, 2005.
16. N. C. Schwertman, M. A. Owens, and R. Adnan, “A simple more general boxplot method for identifying outliers,” Computational Statistics & Data Analysis, vol. 47, no. 1, pp. 165–174, 2004.
17. J. Yang, S. Rahardja, and P. Fränti, “Outlier detection: How to threshold outlier scores?,” in Proc. 2019 International Conference on Artificial Intelligence, Information Processing and Cloud Computing (AIIPCC), Sanya, China, Dec. 19–21, 2019, pp. 1–6.
18. H. Fitriyah and A. S. Budi, “Outlier detection in object counting based on hue and distance transform using median absolute deviation (MAD),” in Proc. 2019 International Conference on Sustainable Information Engineering and Technology (SIET), Malang, Indonesia, Sep. 28–30, 2019, pp. 217–222.
19. S. Mehrang, E. Helander, M. Pavel, A. Chieh, and I. Korhonen, “Outlier detection in weight time series of connected scales,” in Proc. 2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Nov. 9–12, 2015, pp. 1489–1496.
20. D. Anelli, P. Morano, F. Tajani, and M. R. Guarini, “The interpretative effects of normalization techniques on complex regression modeling: An application to real estate values using machine learning,” Information, vol. 16, no. 6, Art. no. 486, 2025.
21. V. Aggarwal, V. Gupta, P. Singh, K. Sharma, and N. Sharma, “Detection of spatial outlier by using improved z-score test,” in Proc. 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI), Apr. 23–25, 2019, pp. 788–790.
22. P. J. Rousseeuw and M. Hubert, “Robust statistics for outlier detection,” WIREs Data Mining and Knowledge Discovery, vol. 1, no. 1, pp. 73–79, 2011.
23. S. Krishna, “How to detect outliers with Z-Score in Python,” Machine Learning Plus, [Online]. Available: https://www.machinelearningplus.com/machine-learning/how-to-detect-outliers-with-z-score/. [Accessed: Apr. 24, 2025].
24. L. Breiman, “Random forests,” Machine Learning, vol. 45, pp. 5–32, 2001.
25. P. F. Smith, S. Ganesh, and P. Liu, “A comparison of random forest regression and multiple linear regression for prediction in neuroscience,” Journal of Neuroscience Methods, vol. 220, no. 1, pp. 85–91, 2013.
26. Ö. G. Uzut and S. Buyrukoğlu, “Veri madenciliği algoritmaları ile gayrimenkul fiyatlarının tahmini,” Euroasia Journal of Mathematics, Engineering, Natural & Medical Sciences., vol. 7, no. 9, pp. 77–84, 2020.
27. Ç. Ersin and M. Yaz, “Comparison of KNN and random forest algorithms in classifying EMG signals,” Avrupa Bilim ve Teknoloji Dergisi, no. 51, pp. 209–216, 2023.
28. B. Gülmez and S. Kulluk, "Analysis and price prediction of secondhand vehicles in Türkiye with big data and machine learning techniques," Journal of the Faculty of Engineering and Architecture of Gazi University, vol. 38, no. 4, pp. 2279–2290, 2023.
29. A. Ç. Aydınoğlu, R. Bovkır, and İ. Çölkesen, “Toplu gayrimenkul değerlemede makine öğrenme algoritmalarının kullanımı ve konumsal/konumsal olmayan özniteliklerin tahmin doğruluğuna etkilerinin karşılaştırılması,” Jeodezi ve Jeoinformasyon Dergisi, vol. 10, no. 1, pp. 63–83, 2023.
30. M. A. Kara, “Makine öğrenmesi algoritmalarıyla konut fiyatlarının tahmini: Giresun örneği,” Journal of Statistics and Applied Sciences, no. 10, pp. 105–111, 2024.
31. S. Şişman and A. Ç. Aydınoğlu, “Toplu gayrimenkul değerlemesi için coğrafi verinin işlenmesi ve rastgele orman makine öğrenmesi tekniğiyle tahmin modelinin geliştirilmesi,” Harita Dergisi, no. 169, pp. 13–27, 2023. “Pendik,” Wikipedia, 2025. [Online]. Available: https://tr.wikipedia.org/wiki/Pendik. [Accessed: Apr. 24, 2025]. Tapu ve Kadastro Genel Müdürlüğü (TKGM), “Parsel Sorgu Uygulaması,” TKGM, [Online]. Available: https://parselsorgu.tkgm.gov.tr/. [Accessed: Apr. 27, 2025].
32. Endeksa, “Gayrimenkul Değerleme ve Pazar Analizi Platformu,” Endeksa, [Online]. Available: https://www.endeksa.com/tr/. [Accessed: Apr. 10, 2025].
33. S. Yalpir, S. Sisman, A. U. Akar, and F. B. Unel, “Feature selection applications and model validation for mass real estate valuation systems,” Land Use Policy, vol. 108, p. 105539, 2021.
34. S. Yalpir, “Enhancement of parcel valuation with adaptive artificial neural network modeling,” Artificial Intelligence Review, vol. 49, no. 3, pp. 393–405, 2018.