Automated Product Recommendations with Preference-Based Explanations

Published on Sep 1, 2020in Journal of Retailing
· DOI :10.1016/J.JRETAI.2020.01.001
André Marchand8
Estimated H-index: 8
(University of Cologne),
Paul Marx3
Estimated H-index: 3
(University of Siegen)
Sources
Abstract
Abstract Many online retailers, such as Amazon, use automated product recommender systems to encourage customer loyalty and cross-sell products. Despite significant improvements to the predictive accuracy of contemporary recommender system algorithms, they remain prone to errors. Erroneous recommendations pose potential threats to online retailers in particular, because they diminish customers’ trust in, acceptance of, satisfaction with, and loyalty to a recommender system. Explanations of the reasoning that lead to recommendations might mitigate these negative effects. That is, a recommendation algorithm ideally would provide both accurate recommendations and explanations of the reasoning for those recommendations. This article proposes a novel method to balance these concurrent objectives. The application of this method, using a combination of content-based and collaborative filtering, to two real-world data sets with more than 100 million product ratings reveals that the proposed method outperforms established recommender approaches in terms of predictive accuracy (more than five percent better than the Netflix Prize winner algorithm according to normalized root mean squared error) and its ability to provide actionable explanations, which is also an ethical requirement of artificial intelligence systems.
📖 Papers frequently viewed together
2010RecSys: Conference on Recommender Systems
3 Authors (Paul Marx, ..., André Marchand)
5 Citations
2008
1 Author (Shanle Ma)
2007
20 Citations
References42
Newest
#1Dokyun Lee (CMU: Carnegie Mellon University)H-Index: 9
#2Kartik Hosanagar (UPenn: University of Pennsylvania)H-Index: 23
Recommender systems appear all across the internet. For e-retailers, this represents an opportunity to get more and niche products before customers’ eyes. However, we find that while implementing r...
34 CitationsSource
#1Shuai Zhang (UNSW: University of New South Wales)H-Index: 69
#1Shuai Zhang (UNSW: University of New South Wales)H-Index: 12
Last. Yi Tay (NTU: Nanyang Technological University)H-Index: 26
view all 4 authors...
With the growing volume of online information, recommender systems have been an effective strategy to overcome information overload. The utility of recommender systems cannot be overstated, given their widespread adoption in many web applications, along with their potential impact to ameliorate many problems related to over-choice. In recent years, deep learning has garnered considerable interest in many research fields such as computer vision and natural language processing, owing not only to s...
724 CitationsSource
#1Quentin André (Ad: INSEAD)H-Index: 2
#2Ziv Carmon (Ad: INSEAD)H-Index: 22
Last. Haiyang Yang (Johns Hopkins University)H-Index: 7
view all 10 authors...
Recent developments in the field of artificial intelligence and data analytics are facilitating the automation of some consumer chores (e.g., in smart homes and in self-driving cars) and allow the emergence of big-data-driven, micro-targeting marketing practices (e.g., personalized content recommendation algorithms). We contend that those developments can generate a tension for marketers, consumers, and policy makers: They can, on the one hand, contribute to consumer well-being by making consume...
43 CitationsSource
#1François A. Carrillat (UTS: University of Technology, Sydney)H-Index: 19
#2Renaud Legoux (HEC Montréal)H-Index: 10
Last. Allègre L. Hadida (University of Cambridge)H-Index: 10
view all 3 authors...
Across the many studies of motion picture box office success, unresolved debates and untested assumptions about the contributing factors persist. Using an accessibility–diagnosticity framework and a meta-analysis of 634 effect sizes from 150 studies, the current article seeks to clarify the relationships of star brand equity and product reviews (from consumers and critics) with box office success. The popularity of stars (market and media appeals) exerts a stronger impact on box office success t...
17 CitationsSource
#1Shuai Zhang (UNSW: University of New South Wales)H-Index: 69
#1Shuai Zhang (UNSW: University of New South Wales)H-Index: 12
Last. Yi Tay (NTU: Nanyang Technological University)H-Index: 26
view all 3 authors...
With the growing volume of online information, recommender systems have been an effective strategy to overcome information overload. The utility of recommender systems cannot be overstated, given their widespread adoption in many web applications, along with their potential impact to ameliorate many problems related to over-choice. In recent years, deep learning has garnered considerable interest in many research fields such as computer vision and natural language processing, owing not only to s...
333 CitationsSource
#1Alexander Bleier (BC: Boston College)H-Index: 6
#1Alexander Bleier (BC: Boston College)H-Index: 3
Last. Maik Eisenbeiss (University of Bremen)H-Index: 8
view all 2 authors...
With the amount of online advertising on a steady rise, generic ads noticeably lose effectiveness. In order to break through the clutter, retailers employ a method called retargeting to tailor their advertisements to individual consumers based on inferred interests and preferences. However, while personalization should generally make ads more appealing, the authors use field data to show that the effectiveness of retargeting considerably hinges on consumers’ trust in a respective retailer. To un...
144 CitationsSource
#1Michel Clement (UHH: University of Hamburg)H-Index: 16
#2Steven Wu (UHH: University of Hamburg)H-Index: 2
Last. Marc Fischer (University of Cologne)H-Index: 11
view all 3 authors...
Abstract High financial risks in production and marketing, the hedonic nature of products, and the global cultural relevance of movies have encouraged a substantial number of researchers to analyze the success drivers of movies. This research provides empirical generalizations in managing the supply and demand of motion pictures. Prior empirical research either ignored the endogeneity of box office and screen allocation or was based on selective samples, ignoring the large amount of smaller movi...
22 CitationsSource
#1Fatih Gedikli (Technical University of Dortmund)H-Index: 11
#2Dietmar Jannach (Technical University of Dortmund)H-Index: 52
Last. Mouzhi Ge (Bundeswehr University Munich)H-Index: 16
view all 3 authors...
Recommender systems help users locate possible items of interest more quickly by filtering and ranking them in a personalized way. Some of these systems provide the end user not only with such a personalized item list but also with an explanation which describes why a specific item is recommended and why the system supposes that the user will like it. Besides helping the user understand the output and rationale of the system, the provision of such explanations can also improve the general accept...
136 CitationsSource
#1Nava Tintarev (Aberd.: University of Aberdeen)H-Index: 16
#2Judith Masthoff (Aberd.: University of Aberdeen)H-Index: 28
When recommender systems present items, these can be accompanied by explanatory information. Such explanations can serve seven aims: effectiveness, satisfaction, transparency, scrutability, trust, persuasiveness, and efficiency. These aims can be incompatible, so any evaluation needs to state which aim is being investigated and use appropriate metrics. This paper focuses particularly on effectiveness (helping users to make good decisions) and its trade-off with satisfaction. It provides an overv...
177 CitationsSource
#1Thorsten Hennig-Thurau (City University London)H-Index: 44
#2André Marchand (WWU: University of Münster)H-Index: 8
Last. Paul Marx (Bauhaus University, Weimar)H-Index: 3
view all 3 authors...
Abstract Because hedonic products consist predominantly of experience attributes, often with many available alternatives, choosing the “right” one is a demanding task for consumers. Decision making becomes even more difficult when a group, instead of an individual consumer, will consume the product, as is regularly the case for hedonic offerings such as movies, opera performances, and wine. Noting the prevalence of automated recommender systems as decision aids, the authors investigate the power...
46 CitationsSource
Cited By9
Newest
#1Shu-Hsien Liao (TKU: Tamkang University)H-Index: 36
#2Retno Widowati (Muhammadiyah University of Yogyakarta)H-Index: 4
Last. Yu-Chieh Hsieh (TKU: Tamkang University)
view all 3 authors...
Abstract null null Online social media create virtual communities and network platforms that people use to create, share, and exchange opinions, views and experiences. With social networks, social commerce not only relies on commerce, but online social media can also promote the sale of goods or services online. Many online operators have begun to use recommendation systems to analyze customer purchase history and identify individual products that customers may purchase. This enables the company...
1 CitationsSource
Source
#1Iqbal H. Sarker (Swinburne University of Technology)H-Index: 14
In the current age of the Fourth Industrial Revolution (4IR or Industry 4.0), the digital world has a wealth of data, such as Internet of Things (IoT) data, cybersecurity data, mobile data, business data, social media data, health data, etc. To intelligently analyze these data and develop the corresponding smart and automated applications, the knowledge of artificial intelligence (AI), particularly, machine learning (ML) is the key. Various types of machine learning algorithms such as supervised...
10 CitationsSource
#1Ming-Hui Huang (NTU: National Taiwan University)H-Index: 21
#2Roland T. Rust (UMD: University of Maryland, College Park)H-Index: 77
Abstract We develop a conceptual framework for collaborative artificial intelligence (AI) in marketing, providing systematic guidance for how human marketers and consumers can team up with AI, which has profound implications for retailing, which is the interface between marketers and consumers. Drawing from the multiple intelligences view that AI advances from mechanical, to thinking, to feeling intelligence (based on how difficult for AI to mimic human intelligences), the framework posits that ...
1 CitationsSource
#1Martin Eling (HSG: University of St. Gallen)H-Index: 31
#2Davide Nuessle (HSG: University of St. Gallen)
Last. Julian Staubli (HSG: University of St. Gallen)
view all 3 authors...
Based on a data set of 91 papers and 22 industry studies, we analyse the impact of artificial intelligence on the insurance sector using Porter’s (1985) value chain and Berliner’s (1982) insurability criteria. Additionally, we present future research directions, from both the academic and practitioner points of view. The results illustrate that both cost efficiencies and new revenue streams can be realised, as the insurance business model will shift from loss compensation to loss prediction and ...
1 CitationsSource
#1Iqbal H. SarkerH-Index: 14
1 CitationsSource
#1Patrick van Esch (AUT: Auckland University of Technology)H-Index: 11
#2Yuanyuan (Gina) Cui (AUT: Auckland University of Technology)H-Index: 4
Last. Shailendra Pratap Jain (UW: University of Washington)H-Index: 14
view all 3 authors...
Artificial intelligence (AI) has penetrated the marketing landscape and is having a profound impact on businesses’ communication strategies. With AI coming under the spotlight, we know surprisingly...
6 CitationsSource
#1Xin Wang (UWO: University of Western Ontario)H-Index: 7
#2Jun Hyun (Joseph) Ryoo (CityU: City University of Hong Kong)
Last. Praveen K. Kopalle (Dartmouth College)H-Index: 30
view all 4 authors...
Abstract This research presents the use of machine learning analytics and metrics in the retailing context. We first discuss what is machine learning and explain the field’s origins. We then demonstrate the strengths of machine learning methods using an online retailing dataset, noting key areas of divergence from the traditional explanatory approach to data analysis. We then provide a review of the current state of machine learning in top-level retailing and marketing research, integrating idea...
1 CitationsSource
Source