Optuna Tabanlı Hiper Parametre Optimizasyonu ile Konut Fiyat Tahminlemede Makine Öğrenmesi Tekniklerinin Karşılaştırmalı Analizi

Öz

Konut fiyatlarının etkili bir şekilde tahmin edilmesi, ekonominin şekillenmesinde kritik bir rol oynamaktadır. Bu çalışmanın amacı, konut fiyatlarını tahminlemede en iyi performans gösteren makine öğrenmesi modelini belirlemektir. Bu amaçla, 10 farklı denetimli regresyon algoritması kullanılarak çeşitli modeller eğitilmiştir. Modellerin performansını optimize etmek amacıyla Grid Search, Random Search ve Optuna gibi hiper parametre ayarlama yöntemleri uygulanmıştır. Eğitim ve test setlerinde elde edilen metrik değerler, modellerin genel performansını değerlendirmek için kullanılmıştır. Araştırma sonuçları, hiper parametre ayarlama yöntemlerinin modellerin genel başarısını etkileyen kritik bir faktör olduğunu göstermiştir. Optuna ile optimize edilen Gradyan Artırma Regresyonu modeli, test veri setinde elde ettiği yüksek R2 değeri (0.6558) ve düşük RMSE değeri (4469.48) ile konut fiyatlarını tahminlemede en başarılı model olarak belirlenmiştir. Optuna, hiper parametre optimizasyonunda sağladığı hassasiyet ve etkinlik ile diğer yöntemlere kıyasla belirgin bir üstünlük sunmuştur.

Anahtar Kelimeler

• Konut fiyat tahmini
• Makine öğrenmesi
• Performans karşılaştırması
• Hiper parametre optimizasyonu
• Optuna

A Comparative Analysis of Machine Learning Techniques for House Price Prediction with Optuna-Based Hyperparameter Optimization

Abstract

Effectively predicting house prices plays a critical role in shaping the economy. This study aims to identify the best-performing machine learning model for predicting house prices. For this purpose, various models were trained using 10 different supervised regression algorithms. Hyperparameter tuning methods such as Grid Search, Random Search, and Optuna were applied to optimize the performance of these models. Metric values obtained from the training and test sets were used to evaluate the overall performance of the models. The research results indicate that hyperparameter tuning methods are a critical factor influencing the overall success of the models. The Gradient Boosting Regressor model optimized with Optuna was identified as the most successful model for predicting house prices, achieving a high R2 score (0.6558) and a low RMSE value (4469.48) on the test dataset. Optuna demonstrated a significant advantage in hyperparameter optimization compared to other methods due to its precision and efficiency.

Keywords

• House price prediction
• Machine learning
• Performance comparison
• Hyperparameter optimization
• Optuna

Kaynakça

1. [1] Sornette D, Woodard R. Financial bubbles, real estate bubbles, derivative bubbles, and the financial and economic crisis. In: Econophysics approaches to large-scale business data and financial crisis. Springer Japan; 2010: 101-148.
2. [2] Tse RY. An application of the ARIMA model to real-estate prices in Hong Kong. Journal of Property Finance. 1997; 8(2): 152-163.
3. [3] Cervero R. Jobs-housing balancing and regional mobility. Journal of the American Planning Association. 1989; 55(2): 136-150.
4. [4] Ghysels E, Plazzi A, Valkanov R, Torous W. Forecasting real estate prices. In: Handbook of Economic Forecasting. Vol. 2. 2013: 509-580.
5. [5] Hutchison NE. Housing as an investment? A comparison of returns from housing with other types of investment. Journal of Property Finance. 1994; 5(2): 47-61.
6. [6] Pai PF, Wang WC. Using machine learning models and actual transaction data for predicting real estate prices. Applied Sciences. 2020; 10(17): 5832.
7. [7] Peter NJ, Okagbue HI, Obasi EC, Akinola AO. Review on the application of artificial neural networks in real estate valuation. International Journal. 2020; 9(3): 5-11.
8. [8] Herath SK, Maier G. The hedonic price method in real estate and housing market research. A review of the literature. Institute for Regional Development and Environment. 2010: 1-21. Vienna, Austria: University of Economics and Business.

9. [9] Taylor LO. Theoretical foundations and empirical developments in hedonic modeling. In: Hedonic methods in housing markets: Pricing environmental amenities and segregation. 2008: 15-37. New York, NY: Springer New York.
10. [10] Landajo M, Bilbao C, Bilbao A. Nonparametric neural network modeling of hedonic prices in the housing market. Empirical Economics. 2012; 42: 987-1009.
11. [11] Northcraft GB, Neale MA. Experts, amateurs, and real estate: An anchoring-and-adjustment perspective on property pricing decisions. Organizational Behavior and Human Decision Processes. 1987; 39(1): 84-97.
12. [12] Patil P. A comparative study of different time series forecasting methods for predicting traffic flow and congestion levels in urban networks. International Journal of Information and Cybersecurity. 2022; 6(1): 1-20.
13. [13] Venkatachalam AR, Sohl JE. An intelligent model selection and forecasting system. Journal of Forecasting. 1999; 18(3): 167-180.
14. [14] Bojer CS. Understanding machine learning-based forecasting methods: A decomposition framework and research opportunities. International Journal of Forecasting. 2022; 38(4): 1555-1561.
15. [15] Almeida JS. Predictive non-linear modeling of complex data by artificial neural networks. Current Opinion in Biotechnology. 2002; 13(1): 72-76.
16. [16] Vanschoren J, Van Rijn JN, Bischl B, Torgo L. OpenML: Networked science in machine learning. ACM SIGKDD Explorations Newsletter. 2014; 15(2): 49-60.
17. [17] L’heureux A, Grolinger K, Elyamany HF, Capretz MA. Machine learning with big data: Challenges and approaches. IEEE Access. 2017; 5: 7776-7797.
18. [18] Felzmann H, Fosch-Villaronga E, Lutz C, Tamò-Larrieux A. Towards transparency by design for artificial intelligence. Science and Engineering Ethics. 2020; 26(6): 3333-3361.
19. [19] Maharana K, Mondal S, Nemade B. A review: Data pre-processing and data augmentation techniques. Global Transitions Proceedings. 2022; 3(1): 91-99.
20. [20] Bejani MM, Ghatee M. A systematic review on overfitting control in shallow and deep neural networks. Artificial Intelligence Review. 2021; 1-48.
21. [21] Janiesch C, Zschech P, Heinrich K. Machine learning and deep learning. Electronic Markets. 2021; 31(3): 685-695.
22. [22] Lu S, Li Z, Qin Z, Yang X, Goh RSM. A hybrid regression technique for house prices prediction. In: 2017 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM). 2017: 319-323. IEEE.
23. [23] Durganjali P, Pujitha MV. House resale price prediction using classification algorithms. In: 2019 International Conference on Smart Structures and Systems (ICSSS). 2019: 1-4. IEEE.
24. [24] Rahadi RA, Wiryono SK, Koesrindartoto DP, Syamwil IB. Factors affecting housing products price in Jakarta metropolitan region. International Journal of Property Sciences. 2016; 6(1).
25. [25] Alfiyatin AN, Febrita RE, Taufiq H, Mahmudy WF. Modeling house price prediction using regression analysis and particle swarm optimization case study: Malang, East Java, Indonesia. International Journal of Advanced Computer Science and Applications. 2017; 8(10).
26. [26] Osmadi A, Kamal EM, Hassan H, Fattah HA. Exploring the elements of housing price in Malaysia. Asian Social Science. 2015; 11(24): 26.
27. [27] Chau KW, Chin TL. A critical review of literature on the hedonic price model. International Journal for Housing Science and Its Applications. 2003; 27(2): 145-165.
28. [28] Ball MJ. Recent empirical work on the determinants of relative house prices. Urban Studies. 1973; 10(2): 213-233.
29. [29] Rodriguez M, Sirmans C. Managing corporate real estate: evidence from the capital markets. Journal of Real Estate Literature. 1996; 4(1): 13-33.
30. [30] Owusu-Manu DG, Edwards DJ, Donkor-Hyiaman KA, Asiedu RO, Hosseini MR, Obiri-Yeboah E. Housing attributes and relative house prices in Ghana. International Journal of Building Pathology and Adaptation. 2019; 37(5): 733-746.
31. [31] Samuel AL. Some studies in machine learning using the game of checkers. IBM Journal of Research and Development. 1959; 3(3): 210-229.
32. [32] Mitchell TM. Machine learning. McGraw Hill. 1997.
33. [33] Hastie T, Tibshirani R, Friedman JH. The elements of statistical learning: data mining, inference, and prediction. Vol. 2. 2009: 1-758. New York: Springer.
34. [34] James G, Witten D, Hastie T, Tibshirani R. An introduction to statistical learning. Vol. 112. 2013: 18. New York: Springer.
35. [35] Xu C, Lu C, Liang X, Gao J, Zheng W, Wang T, Yan S. Multi-loss regularized deep neural network. IEEE Transactions on Circuits and Systems for Video Technology. 2015; 26(12): 2273-2283.
36. [36] Kotsiantis SB, Zaharakis I, Pintelas P. Supervised machine learning: A review of classification techniques. Emerging Artificial Intelligence Applications in Computer Engineering. 2007; 160(1): 3-24.
37. [37] Speelman D. Logistic regression. Corpus Methods for Semantics: Quantitative Studies in Polysemy and Synonymy. 2014; 43: 487-533.
38. [38] Menard S. Coefficients of determination for multiple logistic regression analysis. The American Statistician. 2000; 54(1): 17-24.
39. [39] McDonald GC. Ridge regression. Wiley Interdisciplinary Reviews: Computational Statistics. 2009; 1(1): 93-100.
40. [40] Ranstam J, Cook JA. LASSO regression. Journal of British Surgery. 2018; 105(10): 1348-1348.
41. [41] Zhang F, O'Donnell LJ. Support vector regression. In: Machine learning. 2020: 123-140. Academic Press.
42. [42] Cristianini N, Shawe-Taylor J. An introduction to support vector machines and other kernel-based learning methods. Cambridge University Press. 2000.
43. [43] Xu M, Watanachaturaporn P, Varshney PK, Arora MK. Decision tree regression for soft classification of remote sensing data. Remote Sensing of Environment. 2005; 97(3): 322-336.
44. [44] Li Y, Zou C, Berecibar M, Nanini-Maury E, Chan JCW, Van den Bossche P, et al. Random forest regression for online capacity estimation of lithium-ion batteries. Applied Energy. 2018; 232: 197-210.
45. [45] Fu MC, Qu H. Regression models augmented with direct stochastic gradient estimators. INFORMS Journal on Computing. 2014; 26(3): 484-499.
46. [46] Zhang L, Liu Q, Yang W, Wei N, Dong D. An improved k-nearest neighbor model for short-term traffic flow prediction. Procedia-Social and Behavioral Sciences. 2013; 96: 653-662.
47. [47] Aitkin M, Foxall R. Statistical modelling of artificial neural networks using the multi-layer perceptron. Statistics and Computing. 2003; 13: 227-239.
48. [48] Liu W, Dou Z, Wang W, Liu Y, Zou H, Zhang B, Hou S. Short-term load forecasting based on elastic net improved GMDH and difference degree weighting optimization. Applied Sciences. 2018; 8(9): 1603.
49. [49] Chai T, Draxler RR. Root mean square error (RMSE) or mean absolute error (MAE)?–Arguments against avoiding RMSE in the literature. Geoscientific Model Development. 2014; 7(3): 1247-1250.
50. [50] Prasad NN, Rao JN. The estimation of the mean squared error of small-area estimators. Journal of the American Statistical Association. 1990; 85(409): 163-171.
51. [51] Willmott CJ, Matsuura K. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Research. 2005; 30(1): 79-82.
52. [52] Onyutha C. A hydrological model skill score and revised R-squared. Hydrology Research. 2022; 53(1): 51-64.
53. [53] Ranjan GSK, Verma AK, Radhika S. K-nearest neighbors and grid search cv based real time fault monitoring system for industries. In: 2019 IEEE 5th International Conference for Convergence in Technology (I2CT). 2019: 1-5. IEEE.
54. [54] Hanifi S, Cammarono A, Zare-Behtash H. Advanced hyperparameter optimization of deep learning models for wind power prediction. Renewable Energy. 2024; 221: 119700.
55. [55] Nguyen HP, Liu J, Zio E. A long-term prediction approach based on long short-term memory neural networks with automatic parameter optimization by Tree-structured Parzen Estimator and applied to time-series data of NPP steam generators. Applied Soft Computing. 2020; 89: 106116.
56. [56] Wang L, Xie S, Li T, Fonseca R, Tian Y. Sample-efficient neural architecture search by learning actions for Monte Carlo Tree Search. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2021; 44(9): 5503-5515.
57. [57] Srinivas P, Katarya R. hyOPTXg: OPTUNA hyper-parameter optimization framework for predicting cardiovascular disease using XGBoost. Biomedical Signal Processing and Control. 2022; 73: 103456.
58. [58] Bian K, Priyadarshi R. Machine learning optimization techniques: A survey, classification, challenges, and future research issues. Archives of Computational Methods in Engineering. 2024; 1-25.
59. [59] Zulfiqar M, Gamage KA, Kamran M, Rasheed MB. Hyperparameter optimization of Bayesian neural network using Bayesian optimization and intelligent feature engineering for load forecasting. Sensors. 2022; 22(12): 4446.