The goal of a next basket recommendation (NBR) system is to recommend items for the next basket for a user, based on the sequence of their prior baskets. We examine whether the performance gains of the NBR methods reported in the literature hold up under a fair and comprehensive comparison. To clarify the mixed picture that emerges from our comparison, we provide a novel angle on the evaluation of next basket recommendation (NBR) methods, centered on the distinction between repetition and exploration: the next basket is typically composed of previously consumed items (i.e., repeat items) and new items (i.e., explore items). We propose a set of metrics that measure the repetition/exploration ratio and performance of NBR models. Using these new metrics, we provide a second analysis of state-of-the-art NBR models. The results help to clarify the extent of the actual progress achieved by existing NBR methods as well as the underlying reasons for any improvements that we observe. Overall, our work sheds light on the evaluation problem of NBR, provides a new evaluation protocol, and yields useful insights for the design of models for this task.

Next basket recommendation (NBR) is the task of predicting the next set of items based on a sequence of already purchased baskets. It is a recommendation task that has been widely studied, especially in the context of grocery shopping. In next basket recommendation (NBR), it is useful to distinguish between repeat items, i.e., items that a user has consumed before, and explore items, i.e., items that a user has not consumed before. Most NBR work either ignores this distinction or focuses on repeat items. We formulate the next novel basket recommendation (NNBR) task, i.e., the task of recommending a basket that only consists of novel items, which is valuable for both real-world application and NBR evaluation. We evaluate how existing NBR methods perform on the NNBR task and find that, so far, limited progress has been made w.r.t. the NNBR task. To address the NNBR task, we propose a simple bi-directional transformer basket recommendation model (BTBR), which is focused on directly modeling item-to-item correlations within and across baskets instead of learning complex basket representations. To properly train BTBR, we propose and investigate several masking strategies and training objectives: (i) item-level random masking, (ii) item-level select masking, (iii) basket-level all masking, (iv) basket-level explore masking, and (v) joint masking. In addition, an item-basket swapping strategy is proposed to enrich the item interactions within the same baskets. We conduct extensive experiments on three open datasets with various characteristics. The results demonstrate the effectiveness of BTBR and our masking and swapping strategies for the NNBR task. BTBR with a properly selected masking and swapping strategy can substantially improve NNBR performance.

Modern search and recommendation systems are optimized using logged interaction data. There is increasing societal pressure to enable users of such systems to have some of their data deleted from those systems. This paper focuses on "unlearning" such user data from neighborhood-based recommendation models on sparse, high-dimensional datasets. We present caboose, a custom top-k index for such models, which enables fast and exact deletion of user interactions. We experimentally find that caboose provides competitive index building times, makes sub-second unlearning possible (even for a large index built from one million users and 256 million interactions), and, when integrated into three state-of-the-art next-basket recommendation models, allows users to effectively adjust their predictions to remove sensitive items.

Users of online shopping platforms typically purchase multiple items at a time in the form of a shopping basket. Personalized within-basket recommendation is the task of recommending items to complete an incomplete basket during a shopping session. In contrast to the related task of session-based recommendation, where the goal is to complete an ongoing anonymous session, we have access to the shopping history of the user in within-basket recommendation. Previous studies have shown the superiority of neighborhood-based models for session-based recommendation and the importance of personal history in the grocery shopping domain. But their applicability in within-basket recommendation remains unexplored. We propose PerNIR, a neighborhood-based model that explicitly models the personal history of users for within-basket recommendation in grocery shopping. The main novelty of PerNIR is in modeling the short-term interests of users, which are represented by the current basket, as well as their long-term interest, which is reflected in their purchasing history. In addition to the personal history, user neighbors are used to capture the collaborative purchase behavior. We evaluate PerNIR on two public and proprietary datasets. The experimental results show that it outperforms 10 state-of-the-art competitors with a significant margin, i.e., with gains of more than 12% in terms of hit rate over the second best performing approach. Additionally, we showcase an optimized implementation of our method, which computes recommendations fast enough for real-world production scenarios.

Recommender systems have become an essential instrument to connect people to the items that they need. Online grocery shopping is one scenario where this is very clear. So-called user-centered recommendations take a user as input and suggest items based on the user’s preferences. Such user-centered recommendations have received significant attention and uptake. Instead, we focus on an item-centered recommendation task, again in the grocery-shopping scenario. In the reverse next-period recommendation ( RNPR ) task, we are given an item and have to identify potential users who would like to consume it in the next period. We consider three sub-tasks of the overall reverse next-period recommendation ( RNPR ) task, (i) Expl- RNPR , (ii) Rep- RNPR , and (iii) Mixed- RNPR , where we consider different types of target users, i.e., (i) explore users, who are new to a given item, (ii) repeat users, who previously purchased a given item, and (iii) both explore users and repeat users. To address the Expl- RNPR task, we propose a habit-interest fusion model that employs frequency information to capture the repetition-exploration habits of users and that uses pre-trained item embeddings to model the user’s interests. For the Mixed- RNPR task, we propose a repetition-exploration user ranking algorithm to decouple the repetition and exploration task and investigate the tradeoff between targeting different types of users for a given item. Furthermore, to reduce the computational cost at inference, we analyze the repetition behavior from both user and item perspectives and then introduce a repetition-based candidate filtering method for each sub-task. We conduct experiments on two public grocery-shopping datasets. Our experimental results not only demonstrate the difference between repetition and exploration, but also the effectiveness of the proposed methods.

Next basket recommendation ( NBR) is a special type of sequential recommendation that is increasingly receiving attention. So far, most NBR studies have focused on optimizing the accuracy of the recommendation, whereas optimizing for beyond-accuracy metrics, e.g., item fairness and diversity remains largely unexplored. Recent studies into NBR have found a substantial performance difference between recommending repeat items and explore items. Repeat items contribute most of the users' perceived accuracy compared with explore items. Informed by these findings, we identify a potential "short-cut" to optimize for beyond-accuracy metrics while maintaining high accuracy. To leverage and verify the existence of such short-cuts, we propose a plug-and-play two-step repetition-exploration (TREx) framework that treats repeat items and explores items separately, where we design a simple yet highly effective repetition module to ensure high accuracy, while two exploration modules target optimizing only beyond-accuracy metrics. Experiments are performed on two widely-used datasets w.r.t. a range of beyond-accuracy metrics, viz. five fairness metrics and three diversity metrics. Our experimental results show that: (i) we can achieve state-of-the-art performance w.r.t. accuracy via the designed repetition module in TREx; and (ii) the simple TREx framework achieves "better" beyond-accuracy performance than existing sophisticated methods. Prima facie, this appears to be good news: we can achieve high accuracy and improved beyond-accuracy metrics at the same time. However, we argue that the real-world value of our algorithmic solution, TREx, is likely to be limited and reflect on the reasonableness of the evaluation setup. We end up challenging existing evaluation paradigms, particularly in the context of beyond-accuracy metrics, and provide insights for researchers to navigate potential pitfalls and determine reasonable metrics to consider when optimizing for accuracy and beyond-accuracy metrics.