Explainable Artificial Intelligence for Retail Price Estimation and Pricing-Risk Detection Using Multivariate Product Attributes
DOI:
https://doi.org/10.71426/jcdt.v1.si1.pp19-27Keywords:
Retail price estimation, Explainable Artificial Intelligence (XAI), XGBoost, SHAP, Pricing risk.Abstract
Retail price estimation for heterogeneous products requires models that represent nonlinear attribute interactions without concealing the basis of individual estimates. This study develops an Explainable Artificial Intelligence (XAI) system for reference-price estimation and pricing-risk screening using a real 53,940-record diamond transaction dataset containing carat, cut, colour, clarity, geometric measurements, and observed price. Twenty records with non-positive physical dimensions were removed, leaving 53,920 observations. To reduce analogue leakage, records were grouped by cut, colour, clarity, and 0.05-carat specification buckets before three-fold grouped cross-validation. Decision Tree, Random Forest, Extra Trees, histogram gradient boosting, and XGBoost regressors were evaluated on out-of-fold predictions after logarithmic target transformation. Extra Trees produced the lowest error, with MAE of USD 280.96, RMSE of USD 549.90, MAPE of 7.18%, and R2 = 0.9810. XGBoost achieved RMSE of USD 596.87 and R2 = 0.9776 and was retained as the explanation engine because exact TreeSHAP values are available for tree ensembles. SHAP analysis showed that geometric size variables and carat dominated the learned reference-price function, while clarity and colour provided secondary contributions. A residual-based audit layer translated reference-price deviations into controllable review workloads; a 20% deviation threshold flagged 2,825 records, or 5.24% of the cleaned dataset. The resulting system supports auditable price review while explicitly separating prediction, explanation, and managerial action.
References
[1] Rosen S. Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition. Journal of Political Economy. 1974 Jan 1;82(1):34-55. Available from: https://doi.org/10.1086/260169
[2] Edmonds Jr RG. A Theoretical Basis for Hedonic Regression: A Research Primer. Real Estate Economics. 1984 Mar;12(1):72-85. Available from: https://doi.org/10.1111/1540-6229.00311
[3] Tomkovick C, Dobie KE. Applying Hedonic Pricing Models and Factorial Surveys at Parker Pen to Enhance New Product Success. Journal of Product Innovation Management: An International Publication of the Product Development & Management Association. 1995 Sep;12(4):334-45. Available from: https://doi.org/10.1111/1540-5885.1240334
[4] Roheim CA, Gardiner L, Asche F. Value of Brands and Other Attributes: Hedonic Analysis of Retail Frozen Fish in the UK. Marine Resource Economics. 2007 Jan 1;22(3):239-53. Available from: https://doi.org/10.1086/mre.22.3.42629557
[5] Martínez-Garmendia J. Application of hedonic price modeling to consumer packaged goods using store scanner data. Journal of Business Research. 2010 Jul 1;63(7):690-6. Available from: https://doi.org/10.1016/j.jbusres.2009.05.002
[6] Lee JA, Caudill SB, Mixon Jr FG. Shine bright like a diamond: a hedonic model of grading and pricing an experience good. Applied Economics. 2014 Jun 3;46(16):1829-38. Available from: https://doi.org/10.1080/00036846.2014.884707
[7] Ferreira KJ, Lee BH, Simchi-Levi D. Analytics for an Online Retailer: Demand Forecasting and Price Optimization. Manufacturing & Service Operations Management. 2016 Feb;18(1):69-88. Available from: https://doi.org/10.1287/msom.2015.0561
[8] Kigo SN, Omondi EO, Omolo BO. Assessing predictive performance of supervised machine learning algorithms for a diamond pricing model. Scientific Reports. 2023 Oct 12;13(1):17315. Available from: https://doi.org/10.1038/s41598-023-44326-w
[9] Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016 Aug 13. p. 785-794. Available from: https://doi.org/10.1145/2939672.2939785
[10] Lundberg SM, Erion G, Chen H, DeGrave A, Prutkin JM, Nair B, Katz R, Himmelfarb J, Bansal N, Lee SI. From local explanations to global understanding with explainable AI for trees. Nature Machine Intelligence. 2020 Jan;2(1):56-67. Available from: https://doi.org/10.1038/s42256-019-0138-9
[11] Breiman L. Random Forests. Machine Learning. 2001 Oct;45(1):5-32. Available from: https://doi.org/10.1023/A:1010933404324
[12] Geurts P, Ernst D, Wehenkel L. Extremely randomized trees. Machine Learning. 2006 Apr;63(1):3-42. Available from: https://doi.org/10.1007/s10994-006-6226-1
[13] Friedman JH. Greedy function approximation: A gradient boosting machine. Annals of Statistics. 2001 Oct 1:1189-232. Available from: https://doi.org/10.1214/aos/1013203451
[14] Probst P, Wright MN, Boulesteix AL. Hyperparameters and tuning strategies for random forest. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery. 2019 May;9(3):e1301. Available from: https://doi.org/10.1002/widm.1301
[15] Ribeiro MT, Singh S, Guestrin C. "Why Should I Trust You?": Explaining the Predictions of Any Classifier. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016 Aug 13. p. 1135-1144. Available from: https://doi.org/10.1145/2939672.2939778
[16] Arrieta AB, Díaz-Rodríguez N, Del Ser J, Bennetot A, Tabik S, Barbado A, García S, Gil-López S, Molina D, Benjamins R, Chatila R. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion. 2020 Jun 1;58:82-115. Available from: https://doi.org/10.1016/j.inffus.2019.12.012
[17] Guidotti R, Monreale A, Ruggieri S, Turini F, Giannotti F, Pedreschi D. A Survey of Methods for Explaining Black Box Models. ACM Computing Surveys (CSUR). 2018 Aug 22;51(5):1-42. Available from: https://doi.org/10.1145/3236009
[18] Goldstein A, Kapelner A, Bleich J, Pitkin E. Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation. Journal of Computational and Graphical Statistics. 2015 Jan 2;24(1):44-65. Available from: https://doi.org/10.1080/10618600.2014.907095
[19] Apley DW, Zhu J. Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Models. Journal of the Royal Statistical Society Series B: Statistical Methodology. 2020 Sep;82(4):1059-86. Available from: https://doi.org/10.1111/rssb.12377
[20] Bodria F, Giannotti F, Guidotti R, Naretto F, Pedreschi D, Rinzivillo S. Benchmarking and survey of explanation methods for black box models. Data Mining and Knowledge Discovery. 2023 Sep;37(5):1719-78. Available from: https://doi.org/10.1007/s10618-023-00933-9
[21] 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 Dec 19;30:79-82. Available from: https://doi.org/10.3354/cr030079
[22] Chicco D, Warrens MJ, Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Computer Science. 2021 Jul 5;7:e623. Available from: https://doi.org/10.7717/peerj-cs.623
[23] Nadeau C, Bengio Y. Inference for the Generalization Error. Advances in Neural Information Processing Systems. 1999;12. Available from: https://doi.org/10.1023/A:1024068626366
[24] Varma S, Simon R. Bias in error estimation when using cross-validation for model selection. BMC Bioinformatics. 2006 Feb 23;7(1):91. Available from: https://doi.org/10.1186/1471-2105-7-91
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Saumen Bandyopadhyay (Author)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Authors retain copyright in their published work subject to the journal's publication and licensing arrangements.
Articles published in JMT are distributed under the Creative Commons Attribution–NonCommercial 4.0 International License (CC BY-NC 4.0).
Under this license, readers may share and adapt published material for non-commercial purposes, provided that:
- Appropriate credit is given to the author or authors.
- The original publication source is acknowledged.
- A link to the applicable license is provided.
- Any modifications are clearly indicated.
Commercial reuse is not permitted under CC BY-NC 4.0 without appropriate authorization from the relevant rights holder.
License details are available at:
Creative Commons Attribution–NonCommercial 4.0 International
DOI prefix: 10.71426
