Türk piyasasında araba ikinci el fiyat tahminleri için çeşitli makine öğrenmesi ve derin öğrenme yaklaşımlarının karşılaştırmalı analizi

Öz

Çevresel kaygıların yoğunlaşmasıyla birlikte, ikinci el araç piyasaları, yeni araçların üretimindeki karbon ayak izini azaltma konusunda çevre dostu bir alternatif olarak ön plana çıkmıştır. Ancak, etkili ve doğru fiyat tahmin mekanizmalarının yetersizliği, bu piyasaların büyüme ve verimliliği üzerinde engelleyici bir etkiye sahip olabilir. Bu çalışma, bu sorunu çözme hedefine yönelik olarak, özellikle Türk ikinci el araç piyasası üzerinde durmuştur ve Türkiye genelindeki farklı online pazarlardan derlenen geniş bir veri seti sunmuştur. Bu veri seti, çeşitli araç türleri, özellikleri ve yeniden satış koşulları hakkında geniş kapsamlı bilgiler sağlamaktadır. Çalışmada, ikinci el araç fiyatlarının tahmininde hem klasik makine öğrenmesi yöntemleri -özellikle karar ağaçları- hem de derin öğrenme modelleri kullanılmıştır. Bu karşılaştırmalı analizin amacı, bu metotların yeniden satış fiyatının belirlenmesinde tahmin gücünü ve şeffaflığı nasıl iyileştirebileceğini değerlendirmektir. Karar ağaçlarının daha yüksek performans göstermiş olmasına rağmen, derin öğrenme tekniklerinin de benzer sonuçlar elde ettiği ve bu nedenle daha fazla optimizasyon ve geliştirme potansiyeli taşıdığı tespit edilmiştir. Yeniden satış fiyatlarının doğru bir şekilde tahmin edilmesi, ikinci el araç piyasalarının işleyişini daha verimli hale getirebilir ve potansiyel alıcılar ve satıcılar için daha çekici kılabilir. Ayrıca bu durum, yeni araç talebini önemli ölçüde azaltarak çevresel sürdürülebilirliğe katkıda bulunabilir.

Anahtar Kelimeler

• Yeniden satış fiyatı tahminleme
• tablo verisi
• derin öğrenme
• karar ağaçları

Comparative analysis of various machine learning and deep learning approaches for car resale price prediction in the turkish market

Abstract

With escalating environmental concerns worldwide, the shift towards second-hand car markets has emerged as an eco-friendly alternative to reduce the carbon footprint associated with manufacturing new vehicles. However, the lack of accurate and efficient price prediction mechanisms may impede the growth and efficiency of these markets. This study, focusing on the Turkish second-hand car market, contributes towards addressing this gap by introducing a unique, comprehensive dataset gathered from various online markets across Turkey, thereby offering a broad spectrum of data pertaining to different vehicle types, specifications, and resale conditions. The study employs both classical machine learning methods, specifically decision trees, and deep learning models to predict used car prices. This comparative analysis aims to assess the potential of these methods in improving the predictability and transparency of resale price determination. Despite the superior performance of decision tree models, the study found that deep learning techniques achieved comparable results, indicating their potential for further optimization and enhancement. The accurate prediction of resale prices could streamline the operations of second-hand car markets, increasing their appeal to potential buyers and sellers. This could also contribute to environmental sustainability by significantly reducing the demand for new cars.

Keywords

• Resale price prediction
• tabular data
• deep learning
• decision trees

Kaynakça

1. C. Erdem, İ. Şentürk, A hedonic analysis of used car prices in Turkey. International Journal of Economic Perspectives, 3, 141-149, 2009.
2. E. Liu, J. Li, A. Zheng, H. Liu and T. Jiang, Research on the prediction model of the used car price in view of the pso-gra-bp neural network. Sustainability, 14, 8993, 2022. https://doi.org/10.3390/su14158993.
3. L. Bukvić, J. Pašagić Škrinjar, T. Fratrović and B. Abramović, Price Prediction and Classification of Used-Vehicles Using Supervised Machine Learning. Sustainability, 14, 17034, 2022. https://doi.org/10.339 0/su142417034.
4. Ö. Çelik and U.Ö. Osmanoğlu, Prediction of the prices of second-hand cars. Avrupa Bilim ve Teknoloji Dergisi, 16, 77-83, 2019. https://doi.org/10.31590 /ejosat.542884.
5. K. Samruddhi, RA Kumar, Used car price prediction using k-nearest neighbor based model. Int. J. Innov. Res. Appl. Sci. Eng. (IJIRASE), 4, 629-632, 2020. https://doi.org/10.29027/IJIRASE.v4.i2.2020.629-632.
6. M. Asghar, K. Mehmood, S. Yasin and Z.M. Khan, Used cars price prediction using machine learning with optimal features. Pakistan Journal of Engineering and Technology, 4, 113-119, 2021. https://doi.org/10.518 46/vol4iss2pp113-119.
7. C. Longani, S.P. Potharaju and S. Deore, Price prediction for pre-owned cars using ensemble machine learning techniques. in: M. Rajesh et al. (Eds.), Recent Trends Intensive Comput, IOS Press, pp. 178-187, Netherlands, 2021.
8. S. Shaprapawad, P. Borugadda, and N. Koshika, Car Price Prediction: An Application of Machine Learning. Proceedings of 2023 International Conference on Inventive Computation Technologies (ICICT), pp. 242-248, Raleigh, USA, 2023.

9. Y. Li, Y.Li, and Y. Liu, Research on used car price prediction based on random forest and LightGBM. Proceedings of 2022 IEEE 2nd International Conference on Data Science and Computer Application (ICDSCA), pp. 539-543, Dalian, China, 2022.
10. T. L. Nikmah, R. M., Syafei, R. Muzayanah, A. Salsabila, and A. A. Nurdin, Prediction of Used Car Prices Using K-Nearest Neighbour, Random Forest, and Adaptive Boosting Algorithm. Int. Conf. Optim. Comput. Appl 1, 17-22, 2022.
11. A. A. Alhakamy, A. Alhowaity, A. A. Alatawi and H. Alsaadi, Are Used Cars More Sustainable? Price Prediction Based on Linear Regression. Sustainability, 15, 911, 2023.
12. S. Yılmaz and İ. H. Selvi, Price Prediction Using Web Scraping and Machine Learning Algorithms in the Used Car Market. Sakarya University Journal of Computer and Information Sciences, 6, 140-148, 2023.
13. S. Voß and S. Lessmann, Resale Price Prediction In The Used Car Market, In Tristan Symposium VIII, pp. 1-4, San Pedro de Atacama, Chile, 2013.
14. D. R. Das Adhikary, R. Sahu and S. Pragyna Panda, Prediction of used car prices using machine learning. In Biologically Inspired Techniques in Many Criteria Decision Making: Proceedings of BITMDM, pp. 131-140, Singapore, 2022.
15. Sahibinden, Turkish online marketplace. https://www .sahibinden.com/, Accessed 8 February 2021.
16. M. Joseph, Pytorch tabular: A framework for deep learning with tabular data. arXiv preprint 2021; arXiv:2104.13638.
17. W. Song, C. Shi, Z. Xiao, Z. Duan, Y. Xu, M. Zhang and J. Tang, Autoint: Automatic feature interaction learning via self-attentive neural networks. Proceedings of the 28th ACM international conference on information and knowledge management, pp. 1161-1170, Beijing, China, 2019.
18. J.R. Quinlan, Induction of decision trees. Machine learning, 1, 81-106, 1986. https://doi.org/10.1007/BF0 0116251.
19. L. Breiman, J. Friedman, R. Olshen and C. Stone, Classification and regression trees. CRC Press. Boca Raton, Florida, 1984.
20. L. Breiman, Random Forests. Machine learning, 45, 5-32, 2001. https://doi.org/10.1023/A:1010933404324.
21. T.K. Ho, The random subspace method for constructing decision forests. IEEE transactions on pattern analysis and machine intelligence, 20, 832-844, 1998. https:// doi.org/10.1109/34.709601.
22. T. Chen and C. Guestrin, Xgboost: A scalable tree boosting system. Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pp. 785-794, New York, USA, 2016.
23. P. Geurts, D. Ernst and L. Wehenkel, Extremely randomized trees. Machine learning, 63, 3-42, 2006. https://doi.org/10.1007/s10994-006-6226-1.
24. L. Prokhorenkova, G. Gusev, A. Vorobev, A.V. Dorogush and A. Gulin, CatBoost: unbiased boosting with categorical features. 32nd Conference on Neural Information Processing Systems (NeurIPS), pp. 1-11, Montréal, Canada, 2018.
25. M. Ali, Moez AliPyCaret: An open source, low-code machine learning library in Python, https://www. pycaret.org/, Accessed 1 January 2023.