Revisiting Bundle Recommendation

Abstract

Product bundling is a commonly-used marketing strategy in both offline retailers and online e-commerce systems. Current research on bundle recommendation is limited by: (1) noisy datasets, where bundles are defined by heuristics, e.g., products co-purchased in the same session; and (2) specific tasks, holding unrealistic assumptions, e.g., the availability of bundles for recommendation directly. In this paper, we propose to take a step back and consider the process of bundle recommendation from a holistic user experience perspective. We first construct high-quality bundle datasets with rich meta information, particularly bundle intents, through a carefully designed crowd-sourcing task. We then define a series of tasks that together, support all key steps in a typical bundle recommendation process, from bundle detection, completion, ranking, to explanation and auto-naming. Finally, we conduct extensive experiments and in-depth analysis that demonstrate the challenges of bundle recommendation, arising from the need for capturing complex relations among users, products and bundles, as well as the research opportunities, especially in graph-based neural methods. To sum up, our study delivers new data sources, opens up new research directions, and provides useful guidance for product bundling in real e-commerce platforms. Our datasets are available at GitHub (https://github.com/BundleRec/bundle_recommendation).

Personalized bundle recommendation using preference elicitation and the Choquet integral

“…Bundle recommendation can be defined as the problem of selecting the best group of items from a potentially very large dataset according to some user preferences. This type of recommendation method can involve a series of tasks such as detecting, completing, and ranking bundles, as well as generating bundle explanations and bundle auto-naming (Sun et al, 2022 ). Bundle recommendation systems can exploit different approaches (Sun et al, 2022 ):…”

“…This type of recommendation method can involve a series of tasks such as detecting, completing, and ranking bundles, as well as generating bundle explanations and bundle auto-naming (Sun et al, 2022 ). Bundle recommendation systems can exploit different approaches (Sun et al, 2022 ):…”

“…The task of bundle recommendation consists of generating sets of related products that users tend to consume together in specific situations. A recommendation system based on bundles offers various advantages, ranging from improving user experience through non-trivial complementary items to boosting sales revenue for sellers via cross-selling (Sun et al, 2022 ) or operational efficiency. To address this task, researchers have proposed a wide variety of approaches, ranging from collaborative filtering (Beladev et al, 2016 ) and data mining (Fang et al, 2018 ) to factorization (Cao et al, 2017 ) and graph neural networks (Deng et al, 2020 ).…”

Knowledge Graph Based Recommender for Automatic Playlist Continuation

“…Amidst this landscape lies the specific challenge of playlist continuation, a type of bundle recommendation [19] that entails suggesting item sequences to users. The emergence of GNNs has positioned them as a promising solution for this task, as evidenced in the work [20], which employs GNNs for semantic-rich representation learning, complemented by hard negative sampling.…”

The Distribution of Etiology and Clinical Presentations of Birth Defects: A Preliminary Assessment of China Birth Cohort Study

“…As a crucial technique for addressing information overload and enhancing user experience, recommendation systems have witnessed significant advancements over the past decade and have been widely used in various web applications such as product recommendation [32,49,51,59], video recommendation [39,54,66], news recommendation [55][56][57], music recommendation [27,47] and so on. In the meanwhile, with the development of deep learning, recommendation systems have gone through several stages.…”