SCOPE-Intent: Segment-Conditioned Probability Calibration for E-Commerce Purchase and Repurchase Prediction

Main Article Content

Xiang Li
Qingdao No.58 High School of Shandong Province, Qingdao, China

DOI: https://doi.org/10.70267/icrafm.202617626633

Keywords

purchase-intent prediction, probability calibration, e-commerce, consumer-demand estimation, behavioral segmentation

Abstract

Purchase-intent models can rank prospective buyers well while returning probabilities whose reliability varies across behavioral contexts. We evaluate SCOPE-Intent, a validation-only segment-conditioned calibration layer, on session purchase prediction (UCI Online Shoppers) and customer-anchor 30-day repurchase prediction (UCI Online Retail). The pipeline fits prediction and soft routing on training data, fits all calibration maps on validation data, and evaluates only once on held-out records. Five-seed paired bootstrap analysis finds that SCOPE-Intent improves Brier score versus global temperature scaling on both tasks (Online Shoppers: 0.00068, 95% CI [0.00055, 0.00080]; Online Retail: 0.00040, [0.00032, 0.00049]). The result is not universal: Platt and beta calibration are substantially better on Online Shoppers, and hard or unshrunk segment controls are not stably worse on Online Retail. The evidence supports task- and comparator-conditioned probability calibration only, not marketing uplift, revenue, causal demand effects, or deployment readiness on these two public benchmarks.

Abstract 22 | PDF Downloads 20

References

  • [1]Sakar, C. O. and Kastro, Y., “Real-time prediction of online shoppers’ purchasing intention using multilayer perceptron and LSTM recurrent neural networks,” Neural Computing and Applications, vol. 31 pp. 6893–6908 2019, doi: 10.1007/s00521-018-3523-0.
  • [2]Sakar, C. O. and Kastro, Y., “Online shoppers purchasing intention dataset,” 2018, doi: 10.24432/C5F88Q.
  • [3]Chen, D., Sain, S. L., and Guo, K., “Data mining for the online retail industry: A case study of RFM model-based customer segmentation using data mining,” Journal of Database Marketing and Customer Strategy Management, vol. 19 no. 3 pp. 197–208 2012, doi: 10.1057/dbm.2012.17.
  • [4]Chen, D., “Online retail dataset,” 2015, doi: 10.24432/C5BW33.
  • [5]Platt, J. C., “Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods,” Advances in large margin classifiers, pp. 61–74 1999.
  • [6]Zadrozny, B. and Elkan, C., “Transforming classifier scores into accurate multiclass probability estimates,” Proceedings of the eighth ACM SIGKDD international conference on knowledge discovery and data mining, pp. 694–699 2002, doi: 10.1145/775047.775151.
  • [7]Niculescu-Mizil, A. and Caruana, R., “Predicting good probabilities with supervised learning,” Proceedings of the 22nd international conference on machine learning, pp. 625–632 2005, doi: 10.1145/1102351.1102430.
  • [8]Naeini, M. P., Cooper, G., and Hauskrecht, M., “Obtaining well calibrated probabilities using bayesian binning,” Proceedings of the AAAI conference on artificial intelligence, vol. 29 pp. 2901–2907 2015.
  • [9]Kull, M., Menezes e Silva Filho, T. de, and Flach, P., “Beta calibration: A well-founded and easily implemented improvement on logistic calibration for binary classifiers,” Proceedings of the 20th international conference on artificial intelligence and statistics, vol. 54 pp. 623–631 2017.
  • [10]Guo, C., Pleiss, G., Sun, Y., and Weinberger, K. Q., “On calibration of modern neural networks,” Proceedings of the 34th international conference on machine learning, vol. 70 pp. 1321–1330 2017.
  • [11]Kull, M., Nieto, M. P., Kängsepp, M., Filho, T. S., Song, H., and Flach, P., “Beyond temperature scaling: Obtaining well-calibrated multi-class probabilities with dirichlet calibration,” Advances in neural information processing systems, vol. 32 2019.
  • [12]Wenger, J., Kjellström, H., and Triebel, R., “Non-parametric calibration for classification,” Proceedings of the 23rd international conference on artificial intelligence and statistics, vol. 108 pp. 178–190 2020.
  • [13]Ma, X. and Blaschko, M., “Meta-cal: Well-controlled post-hoc calibration by ranking,” Proceedings of the 38th international conference on machine learning, vol. 139 pp. 7235–7245 2021.
  • [14]Berta, E., Bach, F., and Jordan, M. I., “Classifier calibration with ROC-regularized isotonic regression,” Proceedings of the 27th international conference on artificial intelligence and statistics, vol. 238 pp. 1972–1980 2024.
  • [15]Filho, T. S., Song, H., Perello-Nieto, M., Santos-Rodriguez, R., Kull, M., and Flach, P., “Classifier calibration: A survey on how to assess and improve predicted class probabilities,” Machine Learning, vol. 112 pp. 3211–3260 2023, doi: 10.1007/s10994-023-06336-7.
  • [16]Vaicenavicius, J., Widmann, D., Andersson, C., Lindsten, F., Roll, J., and Schön, T. B., “Evaluating model calibration in classification,” Proceedings of the 22nd international conference on artificial intelligence and statistics, vol. 89 pp. 3459–3467 2019.
  • [17]Kumar, A., Sarawagi, S., and Jain, U., “Trainable calibration measures for neural networks from kernel mean embeddings,” Proceedings of the 35th international conference on machine learning, vol. 80 pp. 2805–2814 2018.
  • [18]Hébert-Johnson, Úrsula, Kim, M. P., Reingold, O., and Rothblum, G. N., “Multicalibration: Calibration for the (computationally-identifiable) masses,” Proceedings of the 35th international conference on machine learning, vol. 80 pp. 1939–1948 2018.
  • [19]Mukhoti, J., Kulharia, V., Sanyal, A., Golodetz, S., Torr, P., and Dokania, P., “Calibrating deep neural networks using focal loss,” Advances in neural information processing systems, vol. 33 2020.
  • [20]Minderer, M., Djolonga, J., Romijnders, R., Hubis, F., Zhai, X., Houlsby, N., Tran, D., and Lucic, M., “Revisiting the calibration of modern neural networks,” Advances in neural information processing systems, vol. 34 pp. 15682–15694 2021.
  • [21]Laird, N. M. and Ware, J. H., “Random-effects models for longitudinal data,” Biometrics, vol. 38 no. 4 pp. 963–974 1982, doi: 10.2307/2529876.
  • [22]Greenland, S., “Principles of multilevel modelling,” International Journal of Epidemiology, vol. 29 no. 1 pp. 158–167 2000, doi: 10.1093/ije/29.1.158.
  • [23]MacQueen, J. B., “Some methods for classification and analysis of multivariate observations,” Proceedings of the fifth berkeley symposium on mathematical statistics and probability, vol. 1 pp. 281–297 1967.
  • [24]Kingma, D. P. and Ba, J., “Adam: A method for stochastic optimization,” International conference on learning representations, 2015.
  • [25]Chen, T. and Guestrin, C., “XGBoost: A scalable tree boosting system,” Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, pp. 785–794 2016, doi: 10.1145/2939672.2939785.
  • [26]Brier, G. W., “Verification of forecasts expressed in terms of probability,” Monthly Weather Review, vol. 78 no. 1 pp. 1–3 1950, doi: 10.1175/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2.
  • [27]Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., and Salakhutdinov, R., “Dropout: A simple way to prevent neural networks from overfitting,” Journal of Machine Learning Research, vol. 15 no. 56 pp. 1929–1958 2014.
  • [28]Murphy, A. H., “A new vector partition of the probability score,” Journal of Applied Meteorology, vol. 12 no. 4 pp. 595–600 1973, doi: 10.1175/1520-0450(1973)012<0595:ANVOTP>2.0.CO;2.

Article Sidebar

Full Text
PDF
Information
Copy Publication Information
Published
Oct 19, 2026
Conference Proceedings Volume
Vol. 17 (2026): Proceedings of the International Conference on Research in Accounting, Finance and Management (ICRAFM 2026)
Section
Articles
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

Li, Xiang. “SCOPE-Intent: Segment-Conditioned Probability Calibration for E-Commerce Purchase and Repurchase Prediction”. Exploring Science Academic Conference Series, vol. 17, Oct. 2026, pp. 626-33, https://doi.org/10.70267/icrafm.202617626633.