{"type":"ScholarlyArticle","title":"A Local Algorithm for Product Return Prediction in E-Commerce","authors":[{"name":"Yada Zhu","slug":"yada-zhu","given":"Yada","family":"Zhu","affiliation":"IBM Research , Yorktown Heights , NY"},{"name":"Jianbo Li","slug":"jianbo-li","given":"Jianbo","family":"Li","affiliation":"Three Bridges Capital , New York , NY"},{"name":"Jingrui He","slug":"jingrui-he","given":"Jingrui","family":"He","affiliation":"Arizona State University , Tempe , AZ"},{"name":"Brian L. Quanz","slug":"brian-l-quanz","given":"Brian L.","family":"Quanz","affiliation":"IBM Research , Yorktown Heights , NY"},{"name":"Ajay A. Deshpande","slug":"ajay-a-deshpande","given":"Ajay A.","family":"Deshpande","affiliation":"IBM Research , Yorktown Heights , NY"}],"abstract":"With the rapid growth of e-tail, the cost to handle returned online orders also increases significantly and has become a major challenge in the e-commerce industry. Accurate prediction of product returns allows e-tailers to prevent problematic transactions in advance. However, the limited existing work for modeling customer online shopping behaviors and predicting their return actions fail to integrate the rich information in the product purchase and return history (e.g., return history, purchase-no-return behavior, and customer/product similarity). Furthermore, the large-scale data sets involved in this problem, typically consisting of millions of customers and tens of thousands of products, also render existing methods inefficient and ineffective at predicting the product returns.\r\n\r\nTo address these problems, in this paper, we propose to use a weighted hybrid graph to represent the rich information in the product purchase and return history, in order to predict product returns. The proposed graph consists of both customer nodes and product nodes, undirected edges reflecting customer return history and customer/product similarity based on their attributes, as well as directed edges discriminating purchase-no-return and no-purchase actions. Based on this representation, we study a random-walk-based local algorithm for predicting product return propensity for each customer, whose computational complexity depends only on the size of the output cluster rather than the entire graph. Such a property makes the proposed local algorithm particularly suitable for processing the large-scale data sets to predict product returns. To test the performance of the proposed techniques, we evaluate the graph model and algorithm on multiple e-commerce data sets, showing improved performance over state-of-the-art methods.","doi":"10.24963/ijcai.2018/517","url":"https://ijcai.org/proceedings/2018/517","pdf":"https://ijcai.org/proceedings/2018/0517.pdf","markdown":"https://ijcai.org/proceedings/2018/517.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/517","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Machine Learning Applications","track":"Main track","number":517,"pages":{"first":3718,"last":3724},"keywords":["Machine Learning: Clustering","Machine Learning: New Problems","Machine Learning Applications: Big data ; Scalability","Machine Learning Applications: Other Applications","Machine Learning: Data Mining","Machine Learning: Learning Preferences or Rankings"],"videos":[],"references":[{"position":1,"raw":"[Andersen et al., 2006] R. Andersen, F. Chung, and K. Lang. Local graph partitioning using pagerank vectors. In 47th Annual IEEE Symposium on Foundations of Computer Sci- ence, pages 475-486, 2006.","title":"Local graph partitioning using pagerank vectors","authors":"R Andersen, F Chung, K Lang","year":"2006","venue":"47th Annual IEEE Symposium on Foundations of Computer Science","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Andersen et al., 2007] R. Andersen, F. Chung, and K. Lang. Local partitioning for directed graphs using pagerank. In Anthony Bonato and Fan R. K. Chung, editors, Interna- tional Workshop on Algorithms and Models for the Web- Graph, volume 4863, pages 166-178. Springer, 2007.","title":"Local partitioning for directed graphs using pagerank","authors":"R Andersen, F Chung, K Lang","year":"2007","venue":"International Workshop on Algorithms and Models for the Web-Graph","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Cassill, 1998] N. L. Cassill. Do customer returns enhance product and shopping experience satisfaction? The In- ternational Review of Retail, Distribution and Customer Research, 8(1):1-13, 1998.","title":"Do customer returns enhance product and shopping experience satisfaction? The International Review of Retail","authors":"N L Cassill","year":"1998","venue":"Distribution and Customer Research","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Ma and Kim, 2016] Jungmok Ma and Harrison M. Kim. Predictive model selection for forecasting product returns. Journal of Mechanical Design, 2016.","title":"Predictive model selection for forecasting product returns","authors":"Kim ; Ma, Jungmok Ma, Harrison M Kim","year":"2016","venue":"Journal of Mechanical Design","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Petersen and Kumar, 2009] J. A. Petersen and V. Kumar. Are product returns a necessary evil? antecedents and con- sequences. Journal of Marketing, 73(3):35-51, 2009.","title":"Are product returns a necessary evil? antecedents and consequences","authors":"Petersen, ; ] J A Kumar, V Petersen, Kumar","year":"2009","venue":"Journal of Marketing","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Pur et al., 2013] S. Pur, E. Stahl, M. Wittmann, G. Wittmann, and S. Weinfurter. Retourenmanagement im online handel-das beste daraus machen. Technical report, Regensburg: ibi research an der universität Regensburg Gmbh, 2013.","title":"Retourenmanagement im online handel-das beste daraus machen","authors":"S Pur, E Stahl, M Wittmann, G Wittmann, S Weinfurter","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Spielman and Teng, 2013] D. Spielman and S. Teng. A lo- cal clustering algorithm for massive graphs and its applica- tion to nearly linear time graph partitioning. SIAM Journal of Computation, 42:1-26, 2013.","title":"A local clustering algorithm for massive graphs and its application to nearly linear time graph partitioning","authors":"D Spielman, S Teng","year":"2013","venue":"SIAM Journal of Computation","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Toktay et al., 2003] L. B. Toktay, W. L. N. Van, and V. D. R. Guide. Forecasting product returns. Business Aspects of Closed Loop Supply Chains, pages 203-220, 2003.","title":"Forecasting product returns","authors":"L B Toktay, W L N Van, V D R Guide","year":"2003","venue":"Business Aspects of Closed Loop Supply Chains","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Toktay, 2004] B. Toktay. Business Aspects of Closed Loop Supply Chains, chapter Forecasting product returns, pages 203-209. Carnegie Mellon University Press, 2004.","title":"Business Aspects of Closed Loop Supply Chains, chapter Forecasting product returns","authors":"B Toktay","year":"2004","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Urbanke et al., 2015] Patrick Urbanke, Johann Kranz, and Lutz Kolbe. Predicting product returns in e-commerce: the contribution of mahalanobis feature extraction? In Proceedings of the 36th International Conference on In- formation Systems (ICIS), 2015.","title":"Predicting product returns in e-commerce: the contribution of mahalanobis feature extraction?","authors":"Patrick Urbanke, Johann Kranz, Lutz Kolbe","year":"2015","venue":"Proceedings of the 36th International Conference on Information Systems (ICIS)","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Wood, 2001] S. L. Wood. Remote purchase environments: the influence of return policy leniency on two-stage decision processes. Journal of Marketing Research, 38(2):157-169, 2001.","title":"Remote purchase environments: the influence of return policy leniency on two-stage decision processes","authors":"S L Wood","year":"2001","venue":"Journal of Marketing Research","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Yang et al., 2017] Hongxia Yang, Yada Zhu, and Jingrui He. Local algorithm for user action prediction towards display ads. In the 23rd ACM SIGKDD International Con- ference on Knowledge Discovery and Data Mining, pages 2091-2099, 2017.","title":"Local algorithm for user action prediction towards display ads","authors":"Hongxia Yang, Yada Zhu, Jingrui He","year":"2017","venue":"the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}