SavedAdd your journal

Journals / Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi / 2026 / Cilt: 42 Sayı: 3

Comparing Global and Segment-Aware Machine Learning Models for Used Car Price Prediction

İkinci El Araç Fiyat Tahmininde Genel ve Segment-Duyarlı Makine Öğrenmesi Modellerinin Karşılaştırılması

Rukiye Nur Kaçmaz(ERCİYES ÜNİVERSİTESİ)0000-0002-3237-9997
Hakan Onay0009-0005-5354-4145
Pages
—
Publication Date
2026-10-14
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The contribution is a controlled empirical comparison rather than a new learning algorithm.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Background: Used car prices are influenced by model year, mileage, fuel type, transmission, engine size, engine power and market position.
📊
Key Findings & Data:Methods: After 167 exact duplicate extras were removed, 15,244 records were evaluated.
💡
Conclusion & Impact:The contribution is a controlled empirical comparison rather than a new learning algorithm.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:ikinci el araç fiyat tahmini · segment duyarlı modelleme · uzmanlar karışımı

Abstract

Background: Used car prices are influenced by model year, mileage, fuel type, transmission, engine size, engine power and market position. This study compares global and segment-aware machine learning approaches for used car price prediction using the CarDekho dataset from Kaggle. Methods: After 167 exact duplicate extras were removed, 15,244 records were evaluated. The primary analysis used ten sealed 80/20 outer splits. Segment topology was selected by three-fold cross-validation within each outer-training set, and fixed CatBoost/XGBoost expert profiles were combined through an XGBoost routing classifier. Global, hard-routing, soft-gating and oracle variants were compared. All target models used log1p prices and a three-fold out-of-fold Duan retransformation correction. Results: Across the full-market splits, global XGBoost achieved R²=0.8523±0.1120, RMSE=349,460±180,834 INR and MAE=104,138±5,274 INR. Soft gating achieved R²=0.8733±0.1007, RMSE=321,661±173,497 INR and MAE=99,196±5,479 INR. Paired Wilcoxon p values were 0.0039 for R² and RMSE and 0.0020 for MAE. In the seed-42 paired bootstrap, the R² and RMSE intervals included zero, whereas the MAE reduction remained supported. The secondary IQR-restricted analysis yielded almost identical mean R² and RMSE for global and soft-gated models, with a lower MAE for soft gating. Conclusion: Segment-aware modeling reduced average full-market error, but the size of the advantage varied across splits. The contribution is a controlled empirical comparison rather than a new learning algorithm.

Özet

Arka plan: İkinci el araç fiyatları; model yılı, kilometre, yakıt türü, şanzıman, motor hacmi, motor gücü ve aracın pazardaki konumu gibi birçok etkene bağlıdır. Bu çalışmada CarDekho veri seti kullanılarak global ve segment duyarlı makine öğrenmesi yaklaşımları karşılaştırılmıştır. Yöntem: Tam yinelenen 167 ek kayıt çıkarıldıktan sonra 15.244 kayıt değerlendirilmiştir. Birincil analizde on adet kapalı %80/%20 dış veri bölümü kullanılmış; segment yapısı yalnızca her dış eğitim kümesindeki üç katlı çapraz doğrulamayla seçilmiştir. Önceden sabitlenen CatBoost/XGBoost uzmanları, XGBoost yönlendirme sınıflandırıcısıyla birleştirilmiş; global, hard routing, soft gating ve oracle yaklaşımları karşılaştırılmıştır. Log1p hedefi için üç katlı out-of-fold Duan geri dönüş düzeltmesi uygulanmıştır. Bulgular: Tam pazar bölümlerinde global XGBoost için R²=0,8523±0,1120, RMSE=349.460±180.834 INR ve MAE=104.138±5.274 INR; soft gating için sırasıyla 0,8733±0,1007, 321.661±173.497 INR ve 99.196±5.479 INR bulunmuştur. Eşleştirilmiş Wilcoxon p değerleri R² ve RMSE için 0,0039, MAE için 0,0020'dir. Seed 42 bootstrap analizinde R² ve RMSE güven aralıkları sıfırı içerirken MAE azalması desteklenmiştir. İkincil IQR analizinde global ve soft-gated modellerin ortalama R² ve RMSE değerleri birbirine çok yakın, soft-gated modelin MAE değeri daha düşük bulunmuştur. Sonuç: Segment duyarlı modelleme tam pazarda ortalama hatayı azaltmıştır; ancak avantajın büyüklüğü veri bölümüne göre değişmektedir. Çalışmanın katkısı yeni bir algoritma değil, kontrollü bir ampirik karşılaştırmadır.

🔖

Cite This Article

Kaçmaz, R. N., Onay, H. (2026). Comparing Global and Segment-Aware Machine Learning Models for Used Car Price Prediction. Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi, Cilt: 42 Sayı: 3. https://doi.org/10.65520/erciyesfen.2001917

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (27)

  1. Pudaruth, S. (2014). Predicting the price of used cars using machine learning techniques. International Journal of Information & Computation Technology, 4(7), 753–764.
  2. Monburinon, N., Chertchom, P., Kaewkiriya, T., Rungpheung, S., Buya, S., & Boonpou, P. (2018). Prediction of prices for used car by using regression models. 2018 5th International Conference on Business and Industrial Research (ICBIR), 115–119. https://doi.org/10.1109/ICBIR.2018.8391177
  3. Gegic, E., Isakovic, B., Keco, D., Masetic, Z., & Kevric, J. (2019). Car price prediction using machine learning techniques. TEM Journal, 8(1), 113–118. https://doi.org/10.18421/TEM81-16
  4. Cui, B., Ye, Z., Zhao, H., Renqing, Z., Meng, L., & Yang, Y. (2022). Used car price prediction based on the iterative framework of XGBoost+LightGBM. Electronics, 11(18), 2932. https://doi.org/10.3390/electronics11182932
  5. Huang, J., Yu, Z., Ning, Z., & Hu, D. (2022). Used car price prediction analysis based on machine learning. In Proceedings of the 2022 International Conference on Artificial Intelligence, Internet and Digital Economy (ICAID 2022) (pp. 356–364). Atlantis Press. https://doi.org/10.2991/978-94-6463-010-7_37
  6. Breiman, L. (2001). Random forests. Machine Learning, 45, 5–32. https://doi.org/10.1023/A:1010933404324
  7. 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
  8. Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785
  9. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., & Gulin, A. (2018). CatBoost: Unbiased boosting with categorical features. Advances in Neural Information Processing Systems, 31, 6638–6648.
  10. Arik, S. Ö., & Pfister, T. (2021). TabNet: Attentive interpretable tabular learning. Proceedings of the AAAI Conference on Artificial Intelligence, 35(8), 6679–6687. https://doi.org/10.1609/aaai.v35i8.16826
  11. Gorishniy, Y., Rubachev, I., Khrulkov, V., & Babenko, A. (2021). Revisiting deep learning models for tabular data. Advances in Neural Information Processing Systems, 34, 18932–18943.
  12. Grinsztajn, L., Oyallon, E., & Varoquaux, G. (2022). Why do tree-based models still outperform deep learning on typical tabular data? Advances in Neural Information Processing Systems, 35, 507–520.
  13. Jacobs, R. A., Jordan, M. I., Nowlan, S. J., & Hinton, G. E. (1991). Adaptive mixtures of local experts. Neural Computation, 3(1), 79–87. https://doi.org/10.1162/neco.1991.3.1.79
  14. Jordan, M. I., & Jacobs, R. A. (1994). Hierarchical mixtures of experts and the EM algorithm. Neural Computation, 6(2), 181–214. https://doi.org/10.1162/neco.1994.6.2.181
  15. Lloyd, S. (1982). Least squares quantization in PCM. IEEE Transactions on Information Theory, 28(2), 129–137. https://doi.org/10.1109/TIT.1982.1056489
  16. Surjandari, I., Dzikri, A., Dhini, A., Laoh, E., Pangesty, K. D., Aurora, P. S., & Ferrouzi, D. (2024). Stacked generalization with sequential-model based optimization for estimating used car valuation in Indonesia. Engineering, Technology & Applied Science Research, 14(5), 17239–17247. https://doi.org/10.48084/etasr.8226
  17. Madhusudhanan, K., Behrens, G., Stubbemann, M., & Schmidt-Thieme, L. (2024). ProbSAINT: Probabilistic tabular regression for used car pricing. 2024 IEEE International Conference on Big Data (BigData), 2179–2187. https://doi.org/10.1109/BigData62323.2024.10825740
  18. Bergmann, S., & Feuerriegel, S. (2025). Machine learning for predicting used car resale prices using granular vehicle equipment information. Expert Systems with Applications, 263, 125640. https://doi.org/10.1016/j.eswa.2024.125640
  19. Fayyaz, I., Ali, G. G. M. N., & Khairunnesa, S. S. (2025). Advanced feature engineering and machine learning techniques for high accurate price prediction of heterogeneous pre-own cars. Vehicles, 7(3), 94. https://doi.org/10.3390/vehicles7030094
  20. Chen, H., Wang, H., Li, J., Yang, T., Wang, H., & Zhang, F. (2026). A research on the residual value of secondhand new energy vehicles based on CatBoost-BiLSTM model. Journal of Big Data, 13, 41. https://doi.org/10.1186/s40537-026-01374-z
  21. Netayawijit, P., Chansanam, W., & Sorn-In, K. (2026). An optimized data partitioning framework for benchmarking ensemble and linear regression models in used-vehicle price prediction. Engineering, Technology & Applied Science Research, 16(2), 32800–32809. https://doi.org/10.48084/etasr.16878
  22. Kaggle. (n.d.). CarDekho Used Car Dataset. https://www.kaggle.com/datasets/manishkr1754/cardekho-used-car-data (accessed July 20, 2026).
  23. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., & Duchesnay, É. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.
  24. Hoerl, A. E., & Kennard, R. W. (1970). Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 12(1), 55–67. https://doi.org/10.1080/00401706.1970.10488634
  25. Tibshirani, R. (1996). Regression shrinkage and selection via the Lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1), 267–288. https://doi.org/10.1111/j.2517-6161.1996.tb02080.x
  26. Vincent, P., Larochelle, H., Bengio, Y., & Manzagol, P.-A. (2008). Extracting and composing robust features with denoising autoencoders. Proceedings of the 25th International Conference on Machine Learning, 1096–1103. https://doi.org/10.1145/1390156.1390294
  27. Duan, N. (1983). Smearing estimate: A nonparametric retransformation method. Journal of the American Statistical Association, 78(383), 605–610. https://doi.org/10.1080/01621459.1983.10478017