SHAP ve LIME Kullanılarak Açıklanabilir XGBoost Modellemesi: Elektrikli Araçlara Ait Gerçek Veri Üzerine Bir Çalışma

Öz

Bu çalışma, açıklanabilir makine öğrenmesi metotları kullanarak elektrikli araç (EA) fiyatlandırmasının belirleyicilerini incelemektedir. Teknik özelliklerden oluşan kapsamlı bir veri seti üzerinde XGBoost modeli kullanılarak yüksek bir tahmin doğruluğu (R² = 0,958) elde edilmiş; modelin karar mantığını ayrıştırmak için SHAP ve LIME yöntemlerinden yararlanılmıştır. Analizlerimiz, otomotiv değerleme ilkelerinde temel bir dönüşümü ortaya koymaktadır: Temel Bileşenler Analizi (PCA) aracılığıyla sentezlenen bütünleşik performans özellikleri birincil fiyat belirleyicisi olarak öne çıkmakta, bunu doğrusal olmayan eşik etkileri sergileyen araç boyutları izlemektedir. Dikkat çekici biçimde, geleneksel ayırt edici unsurların etkisi sınırlı kalmış; bu durum, elektrikli güç aktarma sistemlerinin geleneksel otomotiv hiyerarşilerini yeniden tanımladığını düşündürmektedir. Tamamlayıcı yorumlanabilirlik yöntemleri tutarlı biçimde, EA fiyatlandırmasının geleneksel statü göstergelerinden ziyade mühendislik özünü ödüllendirdiğini; performans paketleri ve boyutsal eşiklerin net bir pazar ayrışması yarattığını göstermektedir. Bu bulgular, üreticilere premium konumlandırma için nicel mühendislik hedefleri sunarken, tüketicilere özellik değerlemesine ilişkin benzeri görülmemiş bir şeffaflık sağlamaktadır. Çalışma, yorumlanabilir makine öğrenmesinin yalnızca fiyatları tahmin etmekle kalmayıp, elektrikli araç devrimini yöneten yeni ekonomik mantığı, teknolojik entegrasyon ve fiziksel oranların tarihsel otomotiv statü sembollerinin önüne geçtiği bir düzeni açığa çıkardığını ortaya koymaktadır.

Anahtar Kelimeler

• Makine Öğrenmesi
• Açıklanabilir Makine Öğrenmesi
• Elektrikli Araçlar
• SHAP
• LIME

Destekleyen Kurum

N/a

Etik Beyan

Bu çalışma için etik kurul onayı gerekmemektedir. Yazar, bu makale ile ilgili olarak herhangi bir kişi, kurum veya kuruluş ile çıkar çatışması bulunmadığını beyan eder.

Teşekkür

N/a

Interpretable XGBoost Modeling Using SHAP and LIME: A Real-World Electric Vehicle Study

Öz

This study investigates the determinants of electric vehicle (EV) pricing through an interpretable machine learning framework. Using XGBoost on a comprehensive dataset of technical specifications, we achieve high predictive accuracy (R²=0.958) while employing SHAP and LIME to deconstruct the model's decision logic. Our analysis reveals a fundamental shift in automotive valuation principles: integrated performance characteristics synthesized via Principal Component Analysis. It emerges as the primary price driver, followed by vehicle dimensions, which exhibit non-linear threshold effects. Notably, traditional differentiators had minimal impact, suggesting that electric powertrains are redefining conventional automotive hierarchies. The complementary interpretability methods consistently demonstrate that EV pricing rewards engineering substance over traditional status markers, with performance bundles and dimensional thresholds creating clear market stratification. These findings provide manufacturers with quantifiable engineering targets for premium positioning and offer consumers unprecedented transparency into feature valuation. The study establishes that interpretable machine learning not only predicts prices but also uncovers the emerging economic logic governing the electric vehicle revolution, where technological integration and physical proportions supersede historical automotive status symbols.

Anahtar Kelimeler

• Machine Learning
• Explainable Machine Learning
• Electric Vehicles
• SHAP
• LIME

Destekleyen Kurum

N/a

Etik Beyan

Ethics committee approval is not required for this study. The author declares that there is no conflict of interest with any person, institution, or organization related to this article.

Teşekkür

N/a

Kaynakça

1. [1] A. Recalde, R. Cajo, W. Velasquez, M. S. Alvarez-Alvarado, "Machine learning and optimization in energy management systems for plug-in hybrid electric vehicles: a comprehensive review", Energies, 17(13), 3059, 2024. https://doi.org/10.3390/en17133059.
2. [2] I. Demirsoy, "Estimating the Intensity of Point Processes on Linear Networks", Ph.D. dissertation, Florida State University, Florida, USA, 2020.
3. [3] A. M. Salih, Z. Raisi-Estabragh, I. B. Galazzo, P. Radeva, S. E. Petersen, K. Lekadir, G. Menegaz, "A Perspective on Explainable Artificial Intelligence Methods: SHAP and LIME", Advanced Intelligent Systems, 7(1), 2400304, 2025. https://doi.org/10.1002/aisy.202400304.
4. [4] C. Y. Zhang, S. Cho, M. Vasarhelyi, "Explainable artificial intelligence (XAI) in auditing", International Journal of Accounting Information Systems, 46(SI), 100572, 2022. https://doi.org/10.1016/j.accinf.2022.100572.
5. [5] M. Chakraborty, “Explainable Artificial Intelligence (XAI): A Perspective”, Lecture Notes in Networks and Systems, M. Chakraborty, S. P. Chakrabarty, A. Penteado, V. E. Balas, Eds., Singapore. Springer, vol 1148, 2025, 47–63. https://doi.org/10.1007/978-981-97-8457-8_5.
6. [6] M. Yildirim, F. Y. Okay, S. Özdemir, "A comparative analysis on the reliability of interpretable machine learning", Pamukkale Üniversitesi Mühendislik Bilimleri Dergisi, 30(4), 494-508, 2023, https://doi.org/10.5505/pajes.2023.49473.
7. [7] C. Van Zyl, X. M. Ye, R. Naidoo, "Harnessing eXplainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of Grad-CAM and SHAP", Applied Energy, 353, 122079, 2024. https://doi.org/10.1016/j.apenergy.2023.122079.
8. [8] A. Gramegna, P. Giudici, "SHAP and LIME: An Evaluation of Discriminative Power in Credit Risk", Frontiers in Artificial Intelligence, 4, 752558, 2021, https://doi.org/10.3389/frai.2021.752558.

9. [9] K. K. Petersen, B. T. Nallapu, I. Demirsoy, R. B. Lipton, E. Grober, J. C. Morris, J. J. Hassenstab, B. A. Gordon, A. Ezzati "Supervised clustering of MRI measures informed by incident cognitive impairment via aggregated survival model SHAP values", Alzheimer's & Dementia, 20(S2), e089947, 2024. https://doi.org/10.1002/alz.089947.
10. [10] I. Demirsoy, "A machine learning approach to determine optimal TOGG sale price in the european market", International Halich Congress on Multidisciplinary Scientific Research-VII, İstanbul, Türkiye, 23-25 January 2024, p. 739-749.
11. [11] I. Demirsoy, A. Karaibrahimoglu, "Identifying drug interactions using machine learning", Advances in Clinical and Experimental Medicine, 32(8), 829-838, 2023, https://doi.org/10.17219/acem/169852.
12. [12] T. Chen, T. He, M. Benesty, V. Khotilovich, Y. Tang, H. Cho, T. Zhou, "Xgboost: extreme gradient boosting", R package version 0.4-2, 1 4, 2015.
13. [13] T. Chen, C. Guestrin, "XGBoost: A Scalable Tree Boosting System", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 2016, 785–794. https://doi.org/10.1145/2939672.2939785.
14. [14] I. T. Jolliffe, "Principal Components in Regression Analysis", Principal Component Analysis, New York, NY, USA, Springer, 1986, chapter 8, pp. 129-155. https://doi.org/10.1007/978-1-4757-1904-8_8.
15. [15] S. M. Lundberg, S.-I. Lee, "A unified approach to interpreting model predictions", Proceedings of the 31st Annual Conference on Neural Information Processing Systems (NIPS), Long Beach, CA, USA, 04-09 December 2017.
16. [16] I. Givisis, D. Kalatzis, C. Christakis, Y. Kiouvrekis, "Comparing Explainable AI Models: SHAP, LIME, and Their Role in Electric Field Strength Prediction over Urban Areas", Electronics, 14(23), 4766, 2025. https://doi.org/10.3390/electronics14234766.
17. [17] M. T. Ribeiro, S. Singh, C. Guestrin, ""Why should i trust you?": Explaining the predictions of any classifier", Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA, 13-17 August 2016, pp. 1135-1144. https://doi.org/10.1145/2939672.2939778.
18. [18] R. C. Team, R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, 2023.
19. [19] S. Guo, B. Zhang, "Revolutionizing the used car market: Predicting prices with XGBoost", Applied and Computational Engineering, 48(1), 173–180, 2024. https://doi.org/10.54254/2755-2721/48/20241349.
20. [20] X. H. Zang, R. N. Abdullah, L. Li, I. A. Hussain, "Leveraging Six Values for Company Performance: Adaptation of Sustainable Business Model Innovation Strategies in Chinese Electric Vehicle Brand Enterprises", World Electric Vehicle Journal, 15(11), 526, 2024, https://doi.org/10.3390/wevj15110526.
21. [21] H. D. Han, S. X. Sun, "Identifying Heterogeneous Willingness to Pay for New Energy Vehicles Attributes: A Discrete Choice Experiment in China", Sustainability, 16(7), 2949, 2024, https://doi.org/10.3390/su16072949.
22. [22] M. K. Hidrue, G. R. Parsons, W. Kempton, M. P. Gardner, "Willingness to pay for electric vehicles and their attributes", Resource and Energy Economics, 33(3), 686-705, 2011. https://doi.org/10.1016/j.reseneeco.2011.02.002.
23. [23] C. D. Higgins, M. Mohamed, M. R. Ferguson, "Size matters: How vehicle body type affects consumer preferences for electric vehicles", Transportation Research Part A: Policy and Practice, 100, 182-201, 2017. https://doi.org/10.1016/j.tra.2017.04.014.
24. [24] Z. W. Liu, X. L. Liu, F. Q. Zhao, "Research on NEV Platform Development Strategies for Automotive Companies", World Electric Vehicle Journal, 12(4), 201, 2021. https://doi.org/10.3390/wevj12040201.
25. [25] B. O. Varga, A. Sagoian, F. Mariasiu, "Prediction of Electric Vehicle Range: A Comprehensive Review of Current Issues and Challenges", Energies, 12(5), 946, 2019. https://doi.org/10.3390/en12050946.
26. [26] M. Herberz, U. J. J. Hahnel, T. Brosch, "Counteracting electric vehicle range concern with a scalable behavioural intervention", Nature Energy, 7(6), 503-510, 2022. https://doi.org/10.1038/s41560-022-01028-3.
27. [27] X. W. Ma, R. Y. Xie, L. X. Guo, S. F. Niu, L. Cheng, R. Q. Hu, "Range anxiety of battery electric vehicles: Quantification and determinants using real-world data", Transportation Research Part D: Transport and Environment, 146, 104837, 2025. https://doi.org/10.1016/j.trd.2025.104837.
28. [28] K. Bayar, "Performance comparison of electric-vehicle drivetrain architectures from a vehicle dynamics perspective", Proceedings of the Institution of Mechanical Engineers Part D: Journal of Automobile Engineering, 234(4), 915-935, 2020. https://doi.org/10.1177/0954407019867491.
29. [29] H. J. Wang, Q. X. Liang, J. T. Hancock, T. M. Khoshgoftaar, "Feature selection strategies: a comparative analysis of SHAP-value and importance-based methods", Journal of Big Data, 11(1), 44, 2024. https://doi.org/10.1186/s40537-024-00905-w.
30. [30] M. Bacevicius, A. Paulauskaite-Taraseviciene, G. Zokaityte, L. Kersys, A. Moleikaityte, "Comparative Analysis of Perturbation Techniques in LIME for Intrusion Detection Enhancement", Machine Learning and Knowledge Extraction, 7(1), 21, 2025. https://doi.org/10.3390/make7010021.
31. [31] E. Figenbaum, "Perspectives on Norway’s supercharged electric vehicle policy", Environmental Innovation and Societal Transitions, 25, 14-34, 2017. https://doi.org/10.1016/j.eist.2016.11.002.
32. [32] Y. Xiao, "Analysis of Factors Influencing Electric Vehicle Sales Based on the Multiple Linear Regression Model", Proceedings of the 2nd International Conference on Innovations in Applied Mathematics, Physics, and Astronomy-IAMPA, Singapore, 2025 pp. 25-30, https://doi.org/10.5220/0013813400004708.