{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T12:18:57Z","timestamp":1767874737044,"version":"3.49.0"},"reference-count":43,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T00:00:00Z","timestamp":1767830400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>E-commerce retailers bear substantial additional costs arising from high product return rates due to lenient return policies and consumers\u2019 impulsive purchasing. This study aims to accurately predict product return behavior before payment, supporting proactive return management and reducing potential losses. Based on the Graph Transformer, we proposed a novel return prediction model, Returnformer, which focuses on capturing user\u2013product connections represented in topological structures of bipartite graphs. The Returnformer first integrates global topological embeddings into original node features to alleviate structural information loss caused by graph partitioning. It then employs a Graph Transformer to capture long-range user\u2013item dependencies within local subgraphs. In addition, a graph-level attention mechanism is introduced to facilitate the propagation of global return patterns across different subgraphs. Experiments on a real-world e-commerce dataset show that the Returnformer outperforms four machine learning models in terms of prediction accuracy, demonstrating superior performance compared to the state-of-the-art models. The proposed model enables retailers to identify potential return risks prior to payment, thereby supporting timely and proactive preventive interventions.<\/jats:p>","DOI":"10.3390\/e28010072","type":"journal-article","created":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T09:01:29Z","timestamp":1767862889000},"page":"72","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Returnformer: A Graph Transformer-Based Model for Predicting Product Returns in E-Commerce"],"prefix":"10.3390","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2602-9472","authenticated-orcid":false,"given":"Qian","family":"Cao","sequence":"first","affiliation":[{"name":"School of Computer and Artificial Intelligence, Beijing Technology and Business University, Beijing 100048, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ning","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer and Artificial Intelligence, Beijing Technology and Business University, Beijing 100048, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huiyong","family":"Li","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence (Institute of Artificial Intelligence), Beihang University, Beijing 100191, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,1,8]]},"reference":[{"key":"ref_1","unstructured":"Statista (2025, September 20). Global Retail e-Commerce Sales 2014\u20132027. Available online: https:\/\/www.statista.com\/statistics\/worldwide-retail-e-commerce-sales."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"176","DOI":"10.1177\/0008125617741125","article-title":"Returning customers: The hidden strategic opportunity of returns management","volume":"60","author":"Huchzermeier","year":"2018","journal-title":"Calif. Manag. Rev."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"345","DOI":"10.1080\/13675567.2017.1370080","article-title":"An exploratory study on the returns management process in an online retailer","volume":"21","author":"Matsuoka","year":"2018","journal-title":"Int. J. Logist. Res. Appl."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"658","DOI":"10.1016\/j.cie.2018.01.007","article-title":"The impact of product returns on price and delivery time competition in online retailing","volume":"125","author":"Zhao","year":"2018","journal-title":"Comput. Ind. Eng."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1125","DOI":"10.1287\/mksc.2023.1451","article-title":"Leveraging the power of images in managing product return rates","volume":"42","author":"Dzyabura","year":"2023","journal-title":"Mark. Sci."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"107811","DOI":"10.1016\/j.resconrec.2024.107811","article-title":"The convenience economy: Product flows and GHG emissions of returned apparel in the EU","volume":"210","author":"Roichman","year":"2024","journal-title":"Resour. Conserv. Recycl."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"143802","DOI":"10.1016\/j.jclepro.2024.143802","article-title":"Return management in e-commerce firms: A machine learning approach to predict product returns and examine variables influencing returns","volume":"477","author":"Mishra","year":"2024","journal-title":"J. Clean. Prod."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"1170","DOI":"10.1016\/j.procs.2020.09.113","article-title":"Predicting the return of orders in the e-tail industry accompanying with model interpretation","volume":"176","author":"Amin","year":"2020","journal-title":"Procedia Comput. Sci."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"495","DOI":"10.1108\/JSIT-06-2020-0120","article-title":"Bayesian-optimized extreme gradient boosting models for classification problems: An experimental analysis of product return case","volume":"26","author":"Bhattacharjee","year":"2024","journal-title":"J. Syst. Inf. Technol."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Li, J., He, J., and Zhu, Y. (2018, January 19\u201323). E-tail product return prediction via hypergraph-based local graph cut. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK.","DOI":"10.1145\/3219819.3219829"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Zhu, Y., Li, J., He, J., Quanz, B.L., and Deshpande, A.A. (2018, January 13\u201319). A local algorithm for product return prediction in e-commerce. Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI), Stockholm, Sweden.","DOI":"10.24963\/ijcai.2018\/517"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"052101","DOI":"10.1007\/s11432-018-9511-1","article-title":"A trust-aware random walk model for return propensity estimation and consumer anomaly scoring in online shopping","volume":"62","author":"Li","year":"2019","journal-title":"Sci. China Inf. Sci."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"McGowan, J., Guest, E., Yan, Z., Zheng, C., Patel, N., Cusack, M., Donaldson, C., de Cnudde, S., Facini, G., and Dzogang, F. (2022, January 18\u201322). A dataset for learning graph representations to predict customer returns in fashion retail. Proceedings of the Workshop on Recommender Systems in Fashion and Retail, Seattle, WA, USA.","DOI":"10.1007\/978-3-031-22192-7_6"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"8060","DOI":"10.3390\/ijerph19138060","article-title":"The impact of information overload of e-commerce platform on consumer return intention: Considering the moderating role of perceived environmental effectiveness","volume":"19","author":"Lv","year":"2022","journal-title":"Int. J. Environ. Res. Public Health"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"114462","DOI":"10.1016\/j.jbusres.2023.114462","article-title":"Online product returns: The role of perceived environmental efficacy and post-purchase entrepreneurial cognitive dissonance","volume":"174","year":"2024","journal-title":"J. Bus. Res."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"3920","DOI":"10.1080\/00207543.2021.1933645","article-title":"Product returns management: A comprehensive review and future research agenda","volume":"60","author":"Ambilkar","year":"2022","journal-title":"Int. J. Prod. Res."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Prayogo, D.H., Domanski, R., and Golinska-Dawson, P. (2024). The key factors for improving returns management in e-commerce in Indonesia from customers\u2019 perspectives\u2014An analytic hierarchy process approach. Sustainability, 16.","DOI":"10.3390\/su16177303"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"451","DOI":"10.1016\/j.ijpe.2013.03.013","article-title":"Examining relationships between the return policy, product quality, and pricing strategy in online direct selling","volume":"144","author":"Li","year":"2013","journal-title":"Int. J. Prod. Econ."},{"key":"ref_19","first-page":"15","article-title":"Improving the marketing strategy into 7PS+ for sustainability: A study of Bali tourism, Indonesia","volume":"104","author":"Yoga","year":"2020","journal-title":"Russ. J. Agric. Socio-Econ. Sci."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1111\/1467-8551.12120","article-title":"Relationship between online retailers\u2019 reputation and product returns","volume":"27","author":"Walsh","year":"2016","journal-title":"Br. J. Manag."},{"key":"ref_21","first-page":"284","article-title":"Uncovering consumers\u2019 returning behaviour: A study of fashion e-commerce","volume":"27","author":"Sutinen","year":"2017","journal-title":"Int. Rev. Retail Distrib. Consum. Res."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"475","DOI":"10.1016\/j.jretai.2024.07.002","article-title":"The faster, the better? The impact of short delivery times on product returns in online retailing","volume":"100","author":"Masuch","year":"2024","journal-title":"J. Retail."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"5058","DOI":"10.1109\/TSMC.2020.2964560","article-title":"Pricing and return strategy: Whether to adopt a cross-channel return option?","volume":"50","author":"Yan","year":"2020","journal-title":"IEEE Trans. Syst. Man Cybern. Syst."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Rajasekaran, V., and Priyadarshini, R. (2021, January 20\u201321). An e-commerce prototype for predicting the product return phenomenon using optimization and regression techniques. Proceedings of the International Conference on Advances in Computing and Data Sciences, Dehradun, India.","DOI":"10.1007\/978-3-030-88244-0_22"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"612","DOI":"10.1016\/j.ejor.2019.05.046","article-title":"Predicting product return volume using machine learning methods","volume":"281","author":"Cui","year":"2020","journal-title":"Eur. J. Oper. Res."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"122755","DOI":"10.1016\/j.techfore.2023.122755","article-title":"Short-lived product returns forecasting when customers are unwilling to return the product: A grey-graphical evaluation and review technique","volume":"195","author":"Tripathy","year":"2023","journal-title":"Technol. Forecast. Soc. Change"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"054501","DOI":"10.1115\/1.4033086","article-title":"Predictive model selection for forecasting product returns","volume":"138","author":"Ma","year":"2016","journal-title":"J. Mech. Des."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1007\/s10479-020-03545-4","article-title":"Policies for inventory models with product returns forecast from past demands and past sales","volume":"288","author":"Chou","year":"2020","journal-title":"Ann. Oper. Res."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1016\/j.dss.2016.08.002","article-title":"Fused latent models for assessing product return propensity in online commerce","volume":"91","author":"Fu","year":"2016","journal-title":"Decis. Support Syst."},{"key":"ref_30","first-page":"353","article-title":"Predicting product returns in e-commerce: The contribution of Mahalanobis feature extraction","volume":"25","author":"Urbanke","year":"2015","journal-title":"Electron. Mark."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"109499","DOI":"10.1016\/j.ijpe.2024.109499","article-title":"Understanding and predicting online product return behavior: An interpretable machine learning approach","volume":"280","author":"Duong","year":"2025","journal-title":"Int. J. Prod. Econ."},{"key":"ref_32","unstructured":"Hofmann, A., Gwinner, F., Fuchs, K., and Winkelmann, A. (2020, January 15\u201317). An industry-agnostic approach for the prediction of return shipments. Proceedings of the Americas Conference on Information Systems (AMCIS), Salt Lake City, UT, USA."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Ma, S., and Wang, W. (2024). Proactive return prediction in online fashion retail using heterogeneous graph neural networks. Electronics, 13.","DOI":"10.3390\/electronics13071398"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Joshi, T., Mukherjee, A., and Ippadi, G. (2018, January 28\u201331). One size does not fit all: Predicting product returns in e-commerce platforms. Proceedings of the 2018 IEEE\/ACM International Conference on Advances in Social Networks Analysis and Mining, Barcelona, Spain.","DOI":"10.1109\/ASONAM.2018.8508486"},{"key":"ref_35","unstructured":"Kedia, S., Madan, M., and Borar, S. (2019). Early bird catches the worm: Predicting returns even before purchase in fashion e-commerce. arXiv."},{"key":"ref_36","unstructured":"Defferrard, M., Bresson, X., and Vandergheynst, P. (2016, January 5\u201310). Convolutional neural networks on graphs with fast localized spectral filtering. Proceedings of the 30th Conference on Neural Information Processing Systems, Barcelona, Spain."},{"key":"ref_37","unstructured":"Kipf, T.N., and Welling, M. (2017, January 24\u201326). Semi-supervised classification with graph convolutional networks. Proceedings of the International Conference on Learning Representations, Toulon, France."},{"key":"ref_38","unstructured":"Veli\u010dkovi\u0107, P., Cucurull, G., Casanova, A., Romero, A., Li\u00f2, P., and Bengio, Y. (May, January 30). Graph attention networks. Proceedings of the International Conference on Learning Representations, Vancouver, BC, Canada."},{"key":"ref_39","unstructured":"Hamilton, W., Ying, Z., and Leskovec, J. (2017, January 4\u20139). Inductive representation learning on large graphs. Proceedings of the 31st Conference on Neural Information Processing Systems, Long Beach, CA, USA."},{"key":"ref_40","unstructured":"Xu, K., Hu, W., Leskovec, J., and Jegelka, S. (May, January 30). How powerful are graph neural networks?. Proceedings of the International Conference on Learning Representations, Vancouver, BC, Canada."},{"key":"ref_41","unstructured":"Dwivedi, V.P., and Bresson, X. (2020). A generalization of transformer networks to graphs. arXiv."},{"key":"ref_42","unstructured":"Liang, J., Chen, M., and Liang, J. (2024, January 21\u201327). Graph external attention enhanced transformer. Proceedings of the International Conference on Machine Learning, Vienna, Austria."},{"key":"ref_43","unstructured":"Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Solja\u010di\u0107, M., Hou, T.Y., and Tegmark, M. (2025, January 22\u201326). KAN: Kolmogorov\u2013Arnold networks. Proceedings of the International Conference on Learning Representations, Singapore."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/28\/1\/72\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T09:07:00Z","timestamp":1767863220000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/28\/1\/72"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,8]]},"references-count":43,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["e28010072"],"URL":"https:\/\/doi.org\/10.3390\/e28010072","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,8]]}}}