Next Basket Recommendation - Potpourri (SOTA)

Next basket recommendation (NBR) aims to infer a set of items that a user will purchase at the next visit by considering a sequence of baskets he/she has purchased previously. Online grocery services can highly benefit from recommender systems, especially when it comes to predicting users’ shopping behavior. This specific scenario owns peculiar characteristics, such as repetitiveness and loyalty, which makes the task very different from the standard recommendations. An efficient solution computes the next basket recommendation, under a more general top-n recommendation framework. A set of collaborative filtering based techniques capture users’ shopping patterns. Recency plays a key role in this particular task. Proposed method [Faggioli 20] is compared with SOTA on two online grocery service datasets.

Human behaviors are complex, which are often observed as a sequence of heterogeneous actions. [Wang 20] takes user choices for shopping baskets as a typical case to study the complexity of user behaviors. Most of existing approaches often model user behaviors in a mechanical way, namely treating a user action sequence as homogeneous sequential data, such as hourly temperatures, which fails to consider the complexity in user behaviors. In fact, users’ choices are driven by certain underlying intentions (e.g., feeding the baby or relieving pain) according to psychological theories. Moreover, the durations of intentions to drive user actions are quite different; some of them may be persistent while others may be transient. A hierarchical framework - Intention Nets (IntNet) - is thus developed to describe the goal, intentions and action sequences. In IntNet, multiple Action Chain Nets are constructed to model the user actions driven by different intentions, and a specially designed Persistent-Transient Intention Unit models the different intention durations.

Existing NBR solutions mainly focus on sequential modeling over their historical interactions. However, due to the diversity and randomness of users’ behaviors, not all these baskets are relevant to help identify the user’s next move. It is necessary to denoise the baskets and extract credibly relevant items to enhance recommendation performance. Unfortunately, this dimension is usually overlooked in the current literature. Contrastive Learning Model (CLEA) [Qin 21] automatically extracts items relevant to target item for next basket recommendation. Specifically, empowered by Gumbel Softmax, a denoising generator is devised to adaptively identify whether each item in a historical basket is relevant to the target item or not. With this process, a positive and negative sub-basket are obtained for each basket over each user. Then, representation of each sub-basket is derived based on its constituent items through a GRU-based context encoder, which expresses either relevant preference or irrelevant noises regarding the target item. After that, a novel two-stage anchor-guided contrastive learning process is then designed to simultaneously guide this relevance learning without requiring any item-level relevance supervision.

Personalized within-basket recommendation [Ariannezhad 23] 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, access to user's shopping history of the user is available in within-basket recommendation. Previous studies have shown superiority of neighborhood-based models for session-based recommendation and importance of personal history in grocery shopping domain. But their applicability in within-basket recommendation remains unexplored. PerNIR, a neighborhood-based model explicitly models personal history of users for within-basket recommendation in grocery shopping. Its main novelty is in modeling short-term interests of users, which are represented by current basket, as well as their long-term interest, which is reflected in their purchasing history. In addition to personal history, user neighbors are used to capture collaborative purchase behavior.

[Faggioli 20] Recency Aware Collaborative Filtering for Next Basket Recommendation

[Wang 20] Intention Nets: Psychology-inspired User Choice Behavior Modeling for Next-basket Prediction

[Qin 21] The world is binary: Contrastive learning for denoising next basket recommendation

[Ariannezhad 23] A Personalized Neighborhood-based Model for Within-basket Recommendation in Grocery Shopping